Skip to main content

Full text of "Social theories of Talcott Parsons: a critical examination"

See other formats






Digitized by the Internet Archive 

in 2011 with funding from 

LYRASIS IVIembers and Sloan Foundation 



















edited by MAX BLACK 

Cornell University 

Englewood Cliffs, N. J. 

Current printing (last digit): 

14 13 12 11 10 9 8 7 6 5 

The Social Theories of Talcott Parsons 



CATALOG CARD NO.: 61-8220 

81961 C 



The following essays have resulted from an intensive study 
at Cornell University of the work of Talcott Parsons. A group 
of ten faculty members, who had been both puzzled and stimu- 
lated by Parsons' writings, met regularly during the academic 
year of 1957-58 for discussion. In the next academic year, there 
followed a series of seven public seminars, widely attended by 
faculty and graduate students, culminating in a session at which 
Parsons himself answered the criticisms he had received. The 
papers here assembled are revised and elaborated versions of 
studies originally prepared for those seminar meetings. 

The warm thanks of all concerned are due to Professor Par- 
sons for the patience, good humor, and generous expenditure 
of time with which he has responded to the labors of his critics. 
Their esteem for him is sufficiently attested by the serious and 
prolonged attention they have given to his investigations. Ac- 
knowledgment is also due to Henry Landsberger, for originat- 
ing the project, and to the Cornell Social Science Research 
Center, and its director, William F. Whyte, without whose sup- 
port little could have been accomplished. 

Choice of one member of the group as editor is a biblio- 
graphical convenience. While all the contributors have learned 
much from one another, each of them is solely responsible for 
the manner and matter of his contribution. 





used in references to M 

Parsons^ writings B 










ASQ A Sociological Approach to the Theory of Organiza- 
tions. Administrative Science Quarterly, I (June 
1956), pp. 63-85; II (September 1956), pp. 225-239. 

Economy and Society (1957) 

Essays in Sociological Theory, first edition (1949) 

Essays in Sociological Theory, revised edition 


Family, Socialization and Interaction Process 


Toward a General Theory of Action (1951) 

The Structure of Social Action (1937) 

The Social System (1951) 

Working Papers in the Theory of Action (1953) 

Note: GTA was published by Harvard University Press, SA by McGraw-Hill, 
the other books by The Free Press, Glencoe, Illinois. A bibliography of Parsons' 
writings through 1959 is included in his Structure and Process in Modern So- 
cieties (Glencoe, III.: The Free Press, 1960). 



= on the Contributors 


ALFRED L. BALDWIN, professor of Child Development and Family 
Relations at Cornell since 1953. Before coming to Cornell as chairman of the 
department he served as professor and chairman of the department of Psy- 
chology at Kansas University, He has also been a Research Associate for the 
Fels Research Institute (1941-49). His pubHcations include Behavior and 
Development in Childhood (1955) and numerous papers. 

MAX BLACK, professor of Philosophy at Cornell since 1946, previouslv 
taught at the University of Illinois and London University. He is a past presi- 
dent of the American Philosophical Association and a member of the Interna- 
tional Institute of Philosophy. His publications include The Nature of 
Mathematics (1933), Critical Thinking (1946, revised ed. 1952), Language 
and Philosophy (1949), and Problems of Analysis (1955). He has been a co- 
editor of The Philosophical Review since 1946. 

URIE BRONFENBRENNER, professor of Child Development and Family 
Relations at Cornell since 1948, has had varied professional experience, in- 
cluding teaching at the University of Michigan and service in the Division of 
Neuropsychiatry, Veterans Administration and in the Office of Strategic Serv- 
ices. His writings include Talent and Society (co-author, 1958), and The 
Measurement of Sociometric Status, Structure, and Development (1945). 

EDWARD C. DEVEREUX, JR., came to Cornell University as professor of 
Child Development and Family Relationships after previous service at Colum- 
bia, Princeton, and the University of Toronto. While a graduate student at 
Harvard, he served for four years as teaching assistant to Talcott Parsons 
( 1936-40 ) . He has been a Fulbright Research Scholar and visiting professor 
at the Institut fiir Sozialforschung in the Goethe University at Frankfurt am 
Main (1956-57). 

ANDREW HACKER has been teaching at Cornell since 1955 in the fields 
of American Government, Political Parties, Political Behavior, and Pohtical 
Theory. In addition to numerous articles in journals, he has written a book 
entitled Political Theory: Philosophy, Ideology, Science (1961). 

HENRY A. LANDSBERGER, associate professor of Industrial and Labor 
Relations, has been assistant director of the Cornell Social Science Research 
Center. He teaches the sociology and the social psychology of industry and is 
the author of Hawthorne Revisited and various papers. His experience includes 
training in clinical psychology and research at the Oxford Institute of 


viii • Notes on the Contributors 

CHANDLER MORSE, professor of Economics, has been at Cornell Uni- 
versity since 1950. He has taught also at Dartmouth and Williams College. 
During the war he held various positions in the Research Branch of the OSS 
and became Assistant Director of Research and Statistics at the Federal Re- 
serve in Washington in 1946. His publications include works on social ac- 
counting, international economics, and the economics of natural resource 
scarcity. A book on the latter subject of which he is co-author is to be pub- 
lished by Resources for the Future, Inc. 

TALCOTT PARSONS, professor of Sociology, Harvard University. He has 
also served as visiting professor of Social Theory at the University of Cam- 
bridge (1953-54) and as a Fellow of the Center for Advanced Study in the 
Behavioral Sciences ( 1957-58 ) . He is a past president of the Eastern Sociolog- 
ical Society and of the American Sociological Association, a Fellow of the 
American Academy of Arts and Sciences and a member of the American 
Philosophical Society. His publications include The Structure of Social Action 
( 1937), Essays in Sociological Theory ( 1949, revised ed. 1954), Toward a Gen- 
eral Theory of Action (with E. A. Shils and others, 1951), The Social System, 
(1951), Economy and Society (with N. J. Smelser, 1956), and Structure and 
Process in Modern Societies (1960). 

WILLIAM FOOTE WHYTE has been a professor of Industrial and Labor 
Relations at Cornell University since 1948 and the director of the Cornell 
Social Science Research Center since 1956. Before coming to Cornell he taught 
at the University of Chicago and the University of Oklahoma. He is the 
editor of Human Organization (official journal of the Society for Applied 
Anthropology) and is a member of the American Academy of Arts and 
Sciences. His writings include the following books: Street Corner Society 
(1943, revised ed. 1955), Human Relations in the Restaurant Industry (1948), 
Pattern for Industrial Peace (1951), Money and Motivation (1955), and Man 
and Organization ( 1959 ) . 

ROBIN M. WILLIAMS, JR., professor of Sociology at Cornell since 1948 
and chairman of the Department of Sociology and Anthropology since 1956. 
He has taught at the University of Kentucky and was Senior Statistical Ana- 
lyst, European Theater of Operations, U.S. War Department, 1943-46. He has 
been a visiting professor at the University of Oslo and at the University of 
Hawaii. He is a past president of the American Sociological Society and of the 
Sociological Research Association. His numerous professional activities include 
service on the Board of Directors of the Social Research Council. Among his 
writings are The Reduction of Intergroup Tensions ( 1947 ) , contributions to 
The American Soldier (1949), American Society: A Sociological Interpretation 
(1951). He is a co-author of Schools in Transition (1954), and What College 
Students Think (1960). 




Edward C. Deveretix, Jr. 

parsons' sociological theory 1 

Robin M. Williams, Jr. 



Chandler Morse 


Alfred L. Baldwin 



Urie Bronfenbrenner 

lRSONs' the 

identification 191 

parsons' THEORY OF 

Henry A. Landsberger 

parsons' theory of ORGANIZATIONS 214 

William Foote Whyte 

parsons' theory applied to 

organizations 250 

Max Black 



Andrew Hacker 


Talcott Parsons 












Edward C. Devereux, Jr. 



In the Dedication o£ The Social System, Talcott Parsons 
describes himself as an incurable theorist. On this one point 
even his severest critics would hasten to agree. Certainly 
he has done a great deal more of theorizing than any other 
contemporary American sociologist; and it is also probably 
true that he has done rather less of anything else. At a 
time when others have been turning more and more to 
empirical research, Parsons has never published a paper 
reporting directly on data derived from a specific empirical 
investigation. And in a generation when others have been 
concerned with "theories of the middle range," Parsons has 
stood virtually alone in his concern with the construction 
of a total, general theoretical system. The magnitude of 
his efforts in pursuit of this single-minded goal is amply 
attested in the long series of theoretical publications listed 
at the end of this volume. 

In this initial paper I shall first say a little about Parsons' 
career, paying particular attention to the sequence of influ- 
ences which seem to have contributed most to the shaping 
of his theory. Second, I shall say something about the theo- 
retical antecedents of Parsonian theory and try to indicate 

2 • Edward C. Devereux, Jr. 

the ways in which Parsons sought to resolve the diflBculties he saw in 
the theories of some of his predecessors. And finally I shall try to 
sketch, with a very broad brush, the main outlines of the theory 
itself, as it has been developing over the past three decades. 

This latter task in particular I approach with some trepidation, 
for as everyone knows the Parsonian theoretical forest is vast and 
tangled, a veritable jungle of fine distinctions and intertwining 
classifications. Moreover it is still growing at a prodigious rate, 
as evidenced by the publication of no less than fourteen additional 
papers in the year after this paper was originally prepared. And, 
like Birnam Wood, it moves: Parsofiian theory in the late 1950's 
differs in some important respects from that of a decade ago. 

Space and the purpose at hand preclude any attempt here to 
examine this Parsonian forest tree by tree and branch by branch. 
My objective is merely to achieve some overall perspective. And 
this, it seems to me, may best be served by standing rather far 
away and squinting a little. 

Let me concede at once that the approach I propose to follow 
here is dangerous. I shall need to be drastically selective in my 
choice of the themes to be discussed, and any such selection must 
inevitably involve at least an element of personal prejudice. Other 
reviewers would undoubtedly select somewhat different points for 
emphasis; and Parsons himself is not much help because of his 
exasperating tendency to insist that each and every point in his 
entire system is fundamental. 

In the interests of communication, moreover, I shall deliberately 
seek to avoid becoming entangled in the peculiar subtleties and ob- 
scurities of Parsonian language. Parsons has been explaining his 
own theories in his own words these many years, but the evidence 
is rather impressive that he has not always succeeded in making 
himself understood. Here, I shall not undertake to explain Parsons 
by quoting him. At the risk of seeming "unscholarly" I shall try to 
state as directly and simply as possible what it is I think he has 
been saying. 

It is not only his language which has placed a barrier between 
Parsons and his readers. There is also his practice of writing at a 
level of sustained abstraction, pyramiding argument upon argu- 
ment with hardly any reference to the realms of empirical phe- 


nomena to which they might conceivably apply. Here, I shall make 
an eflFort to supply empirical referents for some of the Parsonian 
notions under consideration, but let it be clear that at many points 
I have simply had to guess. 

My objective in setting forth this oversimplified account is not, 
however, to provide a primer for students of Parsonian theory. My 
main purpose is rather to call attention as forcibly as I can to the 
fact that Parsonian theory, for all of its intricate complexities and 
details, is primarily a general theory. I am convinced that Parsons 
himself is far more interested in the grand design than he is in any 
particular details. Inevitably, in the course of the development and 
elaboration of his theory, he has developed innumerable detailed 
classifications and attempted innumerable empirical generaliza- 
tions. Inevitably, also, much of the response to his work has been 
focused on these details. As some of the later papers in this book 
bear ample witness, critics have rightfully taken issue with this 
or that particular classification, or have challenged particular em- 
pirical generalizations, or have voiced a general exasperation at the 
fuzziness of Parsonian definitions. Similarly, critics and supporters 
alike have usually managed to find somewhere in Parsons at least 
a few valued nuggets of theoretical construction or empirical in- 

But these are not the grounds, as I see it, on which Parsons 
would prefer to be judged. It is not enough, he would argue, to 
create particular ad hoc classifications, however useful they may 
be, or to hit upon fruitful empirical insights. The main point is 
that these classifications and insights should occur within the frame- 
work of a systematic general theory, should flow from it or some- 
how be generated by it. In the present paper, therefore, I shall try 
to keep the focus directly upon the general theory itself, omitting 
or simplifying the details, but stressing always the over-all plan. 


IN parsons' career 

I was asked to say something about the influences that have 
contributed to the development of Parsons' thinking, and I am in- 

4 • Edward C. Devereux, Jr. 

clined to do so rather briefly. For it seems to me that when one is 
discussing a scientific theory, it is its substance rather than its origin 
that ought to be the main focus of attention. 

With respect to Parsonian theory, however, the question of saH- 
ent exposures may have some special interest. For Parsons has 
always regarded himself as something of a maverick in the field 
of sociology and has exasperated many of his critics by his failure 
to build on the work of other pioneers in American sociology. In- 
deed reviewers have noted with alarm that in Parsons' first full- 
length monograph. The Structure of Social Action, there is not a 
single indexed reference to such figures as Cooley, Ross, W. I. 
Thomas, or G. H. Mead. Clearly Parsons came to sociology by a 
different and devious route. 

My task of reviewing the influences which contributed to Par- 
sons' thinking is greatly lightened by the fact that he has himself 
just written a more extensive account of these influences, to which 
readers wishing more detail are referred.* 

Very sketchily then, let it be noted that Parsons was the son of 
a Congregational clergyman, later to become president of Marietta 
College. He received his undergraduate training at Amherst Col- 
lege, where no sociology has ever been taught. His undergraduate 
major was in biology and at the time he was contemplating a 
career in medicine. Although this particular path was short- 
circuited, Parsons' interests in both biology and medicine have 
survived to play a major role in this thinking. More directly in the 
realm of social science was a philosophy course with Clarence Ayers 
on "The Moral Order" in which Parsons was exposed to the works 
of Sumner, Cooley, and Durkheim, and work in economics with 
Walton Hamilton, the institutional economist, from whom Parsons 
gained an acquaintance with the works of Veblen and John R. 
Commons, and a lifelong interest in the sociological parameters of 
economic activities. 

It was this latter interest which sent Parsons next for a year to 
the London School of Economics, where he studied with L. T. Hob- 
house, Morris Ginsberg, and Bronislaw Malinowski, the first an au- 

* Talcott Parsons, "A Short Account of My Intellectual Development,'" 
Alpha Kappa Deltan, XXIX: 1 (Winter 1959), pp. 3-12. My account 
here leans heavily on this source. 


thority on the evolution of morality, the second an expert on the 
economic institutions of preUterate societies, and the third a pio- 
neer in the development of structural-functional analysis in an- 
thropology. All of these interests, like those aroused at Amherst, 
have remained central in Parsons' thinking throughout his career. 

There followed a year at Heidelberg where Parsons made his 
initial acquaintance with German sociology. The main influences of 
this period stemmed not from living teachers but from the pub- 
lished works of such giants of preceding generations as Max 
Weber, Werner Sombart, and Karl Marx, After a year of teaching 
at Amherst, Parsons returned to Heidelberg to take liis doctoral 
degree. His thesis was concerned with the conceptions of capitalism 
in the literature of German social science, especially in the works 
of the three theorists mentioned. Through his own later work, Par- 
sons was to play a major role in introducing Weber to American 
sociologists. Unfortunately for his American readers. Parsons also 
brought back from Germany the complex, ponderous style of writ- 
ing which has characterized so much of his scholarly output. 

The balance of Parsons' professional career, except for occasional 
leaves of absence, has been spent at Harvard University, first 
briefly as an instructor in the Economics Department, next as an 
instructor and charter member of the newly organized Department 
of Sociology, and finally, after he had become a full professor, as 
founder and first chairman of the Department of Social Relations. 
The influences which have played upon Parsons during these three 
decades are numerous and complex, but perhaps it is fair to say 
only a minor role was played by established sociologists who were 
senior to Parsons, either at Harvard or elsewhere in America. 

During his short period in the Economics Department, he worked 
particularly with F. W. Taussig, to whom he attributes his initial 
and lasting interest in Alfred Marshall, and with T. N. Carver, who 
further sharpened his interest in the moral and ethical problems of 
an industrial society. Parsons' discovery of Vilfredo Pareto, the 
Italian economist-turned-sociologist, led him inevitably into close 
contact with Harvard's famous biochemist, L. J. Henderson, who 
had become a leading disciple of Pareto and who shared and 
fostered Parsons' interest in the parallels between organisms and 
societies as systems. Henderson's book on The Fitness of the En- 

6 • Edward C. Devereux, Jr. 

vironment,'^ together with that o£ Walter Cannon, on The Wisdom 
of the Body f may be counted as major influences in the shaping 
of Parsons' own notions about the properties of systems. Through 
Henderson also, Parsons was drawn into close contact with Elton 
Mayo and others at the Harvard Business School who were then 
engaged in pioneer research and theorizing about the social and 
human problems of industrial organization. 

Although Parsons had never taken any formal course work in 
psychology, his systematic interests had led him to read widely in 
this field. Wolfgang Kohler's works on Gestalt psychology con- 
tributed much to Parsons' thinking about the organized and func- 
tional properties of orientation and perception in goal-seeking situ- 
ations, and E. C. Tolman's pioneer work on purposive behavior in 
animals and man helped to sharpen Parsons' thinking about the 
physiological and psychological roots of goal-seeking behavior. But 
these psychologies seemed relatively thin with respect to the prob- 
lems of personality organization and development, and in these 
areas Parsons was influenced most heavily by Freud and his fol- 
lowers. In addition to his extensive readings in this area. Parsons 
furthered his training by undergoing a didactic analysis. 

I have focused thus far on influences which flowed toward Par- 
sons from others with established positions. In fact, of course, as 
soon as Parsons had established his own solid footing in the field, 
the influences flowed in both directions. This was certainly the case 
with respect to Parsons' interactions with such colleagues and con- 
temporaries as Clyde Kluckhohn in anthropology, O. H. Taylor in 
economics, or Samuel Stouffer in Social Relations. And it was still 
more the case in Parsons' relations with several generations of 
junior colleagues and graduate students who sojourned at Harvard, 
a changing group which in the first decade included such people 
as R. K. Merton, Kingsley Davis, Robin Williams, and Wilbert 
Moore, somewhat later included people like Marion J. Levy, Albert 
Cohen, David Aberle and Bernard Barber, and in the most recent 
decade, R. F. Bales, Edward Shils, James Olds, Renee Fox, and Niel 
Smelser, to mention only a few. Whereas it is true that all of these 

* (New York: The Macmillan Company, 1913.) 
f (New York: Norton, 1932.) 


people were deeply influenced by their contacts with Parsons, it is 
also true that Parsons' own theories have been influenced by his 
interactions with them. 

If it is true that one absorbs a part of all that he has met, we 
should not be surprised to find that various strands of Parsonian 
theory reflect and incorporate elements of biology and medicine, 
of economics, especially of institutional economics, and of the 
utilitarian tradition from which it emerged, of German formal 
sociology, with its propensities for ponderous systematic analysis, 
together with its traditions of idealism and Verstehen, of structural- 
functional analysis as developed by Durkheim and the anthro- 
pologists, and of Gestalt and Freudian psychology. But Parsonian 
theory is not simply an eclectic amalgam of elements drawn from 
these sources. The main point is how its author responded to these 
influences and constructed from them a single systematic theory 
which is uniquely Parsonian. Let us turn at once to the substance 
of the matter. 


The theory of social action which has come to be associated with 
Parsons' name did not spring full-armed from the Parsonian fore- 
head. Before settling down to the serious business of developing 
his own theoretical system. Parsons devoted a decade of produc- 
tive scholarship to a careful critique of the theoretical systems of 
some of his predecessors.* By examining the kinds of substantive 
and methodological issues which seemed most salient for him in 
the works of others, we may hope to learn something of the basic 
orientations which underlie Parsons' own approach to social theory. 

Social theory, as Parsons saw it, had been developing in essen- 
tially three different schools or traditions, each committed to ap- 
parently conflicting notions about the nature of man, societ)' and 

* I cannot take space here to review Parsons' critique of particular 
theorists. The fruits of these critical efforts are set forth in the fifteen 
papers he pubhshed between 1928 and 1937, and in his first book- 
length monograph, SA, pubhshed in 1937. For detailed references, 
see the Bibliography mentioned on page vi above. 

8 • Edward C. Devereux, Jr. 

human behavior, indeed, even of scientific method. The utihtarians 
and classical economists had been attempting to develop essentially 
a rationalistic, individualistic theory of social behavior; the posi- 
tivists had been attempting to develop a theory M^hich could handle 
human behavior in terms of determinate scientific laws of the sort 
which served so well in physics or biology; and the idealists had 
been attempting to develop a theory which interpreted concrete 
social phenomena essentially as emanations from the realm of 
cultural values. 

In Parsons' view, each of these schools had grasped an essential 
part of the truth, but not all of it.. And correspondingly, whereas 
each had successfully developed various special theories, no one by 
itself could provide an adequate basis for a general theory of social 
action. Parsons saw his own task, initially, as one of reconciliation 
and integration. Indeed, he argued that this task was already half 
done before he turned his own attention to it. For it appeared to 
him that in fact certain of the more sophisticated thinkers in each 
of these traditions had already become aware of the limitations of 
their own starting points, and in attempting to overcome them, had 
moved independently toward convergence on a common theoretical 
scheme. In The Structure of Social Action, Parsons attempted to 
document this thesis in some detail with respect to the works of 
Marshall, Durkheim, Pareto, and Max Weber, and to state in more 
general analytical terms the nature of the theoretical framework 
toward which these theorists seemed to be converging. It was this 
framework, designated then as the voluntaristic theory of action, 
which became the core of Parsons' own theory. Virtually all of his 
subsequent work has been devoted to its systematic development 
and elaboration. 

Let us review the issues which seemed most salient for Parsons 
in his analysis of the assets and liabilities of these three apparently 
conflicting approaches to social theory. 

Utilitarianism and Economic Theory 

Economic theory has always had a special appeal for Parsons, 
essentially for two reasons. It is elegantly analytical and systematic, 
and thus represents for Parsons a model of what a really good 


theory ought to look hke. And it is definitely a theory of action: 
the mainspring or force which is assumed ultimately to determine 
the flow of goods and services through the market is always some 
notion of individuals, or firms, orienting themselves in a situation 
and acting in some way designed to advance their economic inter- 
ests. The subjective processes of orientation and decision-making 
and some notion of purposive action are thus integral to the struc- 
ture of the theory. This, Parsons maintains, makes it a very different 
sort of theory from what one would have if economic laws were 
simply statements of statistical uniformities in observed economic 
behavior. It is also basically a better sort of theory, in Parsons' view, 
because it deals with the human qualities of action, directly ac- 
cessible to experience and "understanding," and because it supplies 
a motivational dynamic which helps not only to "explain" observed 
behavior but also to anticipate what might be expected to happen 
under various assumed conditions. Parsons' own objective has al- 
ways been to build a general theory of sociology which would be 
based on some such action principle. 

But Parsons saw economic theory as caught in a difficult dilemma. 
On the one hand, it could hold itself to the task of building tlieo- 
retically neat analytical models, attempting to work through sys- 
tematically what might be expected to occur under various sets 
of carefully but rather arbitrarily and narrowly defined assumptions 
about the nature of economic motivation, rationahty, knowledge, 
competition, rules of the game, and so on. One might point to re- 
cent developments in the theory of games as a rather sophisticated 
exercise in this sort of model building. But economists who hold to 
this conception of their role, Parsons reasoned, must forever accept 
the fact that their elegant systems do not apply very precisely to 
the empirical world of on-going economic activities. For these are 
never purely economic or purely rational, but are always embedded 
in a complex matrix of noneconomic or, broadly, sociological factors. 
To the extent that these noneconomic parameters of economic be- 
havior interact in any fundamental way with the operation of the 
economic system itself, it seemed to Parsons that a purely economic 
theory could never achieve the status of a general theory, even of 
economic behavior. 

On the other hand, economists could attempt to bring their 

10 • Edward C. Devereux, Jr. 

models into closer approximation to empirically observed economic 
behavior by modifying and enlarging the range o£ variables they 
dealt with, to include some reference to the noneconomic or soci- 
ological parameters of their systems. In his review of the literature, 
it seemed to Parsons that most economists had elected this second 
alternative, though not always intentionally. Thus Parsons found 
himself especially interested in Adam Smith's treatment of "moral 
sentiments," in Ricardo's "habits and customs of the people," in 
Marshall's "wants adjusted to activities," in Veblen's "instinct of 
workmanship," in Sombart's "spirit of capitalism," in Tonnies' 
"Gemeinschaft and Gesellschaft," and so on. But while these the- 
ories achieved at least an appearance of sociological sophistication, 
they were certainly more complex, less elegant, and fuzzier than 
those which clung to the strictly economic line. 

In many cases, it appeared to Parsons, sociological considerations 
tended to be treated residually, in a crude, common-sense fashion, 
and were not really incorporated as integral components of the 
theoretical system itself.* Where this was the case, reference to 
these noneconomic factors did not really help toward the establish- 
ment of an adequate general theory. On the other hand. Parsons 
observed, when an attempt was made explicitly to incorporate such 
elements into the body of the theory itself, the result was usually 
to destroy the status of the theory as an action theory and to con- 
vert it to some form of radical, positivistic theory. 

And thus it appeared to Parsons that economic theories either 
left things out which ought to be taken into account, or took them 
into account unsystematically, or built them in in such a way as to 
corrupt the status of the theory as an action theory. These general 
criticisms, moreover, he found also applicable to the utilitarians in 
their eflForts to build a general theory of social behavior on a similar 

* One might observe in passing that until quite recently this has been 
the standard practice of businessmen in dealing with the sociological 
parameters of business decisions. After being briefed by his staff 
of experts on the technical and economic factors in the situation, the 
executive brings the priceless ingredient of "sound judgment" to bear 
upon the "intangibles" in the situation. Although many wise decisions 
have undoubtedly been made in this fashion, they cannot contribute 
much to the advance of science until the variables and principles 
taken into account are given expHcit codification. 

parsons' sociological theory • 11 

rationalistic, action basis. Let us consider briefly the substantive 
issues which Parsons held to be the major sources of difficulty. 

First, there was the problem of order. Most orthodox economic 
theorists tended to ignore this problem and to take order pretty 
much for granted. From Adam Smith right tlirough to T. N. 
Carver, the view was widespread that, given freedom, rationality, 
and enlightened self-interest, people would automatically develop 
systems of cooperation, contract, and exchange which would re- 
sult in mutual benefit for all. It was assumed that the natural 
mechanisms of the market place would somehow create an order 
within which an optimum of wealth and satisfaction could be 

To Parsons, this argument seemed to sidestep the classical 
Hobbesian dilemma: if men are free to pursue their own self-inter- 
est, the paths of rationality will not always lie in the direction of 
cooperation and exchange; for collusion, force, and fraud will also 
present themselves as rationally attractive alternatives. But Hobbes' 
solution, which consisted of invoking a sovereign as a sort of deus 
ex machina to hold a monopoly on the use of force and to use it to 
insure the fulfillment of contractual obligations, seemed to Parsons 
far too pat. For he was convinced that an externally imposed order 
would be extremely precarious and brittle: if there cannot be a 
policeman at every stoplight and every store front, there must be 
some positive motivation to conformity over and beyond the fear 
of sanctions. Locke's solution of the same dilemma, which turned 
upon an implicitly postulated "natural identity of interests," seemed 
to Parsons equally unsatisfactory. He felt that in point of fact, Locke 
was probably right: that social order does indeed depend upon a 
broad stratum of common values and interests. But to take these 
for granted, as somehow given in nature, seemed to beg such es- 
sential questions as where they come from, how they are generated 
and maintained, what determines their particular content, and how 
these diflFerences in content affect the operation of the social system. 

In the end, Parsons concluded that neither the utilitarians nor 
the classical economists were able to develop an adequate general 
theory of social order within the individualistic, rationalistic frame- 
work of their theoretical systems. For order, as he saw it, could not 
be a resultant either of rational self-interest or of externally imposed 

12 • Edward C. Devereux, Jr. 

sanctions alone, but must rest on a core of institutionalized com- 
mon values. The order of the market place, he argued, was es- 
sentially a factual order. If it was also a moral order, it must be 
because of the operation of factors not adequately conceptuaHzed 
in the existing theories. Durkheim's recognition of the noncon- 
tractual element in contracts seemed to Parsons a giant step in the 
right direction. 

On yet a second front, Parsons argued, the preoccupation of the 
utilitarians and orthodox economists with rationality got them into 
serious theoretical difficulties. So long as these theorists concerned 
themselves with rational behavior alone, their action frame of refer- 
ence seemed to work well enough. But as we have seen, many 
theorists were unwilling to let well enough alone and tried in one 
way or another to take into account the fact that not all behavior 
is rational. Given their rationalistic premises, however, it was hard 
for them to see that there could be any mode of departure from 
rationality save that implied in the notion of "irrationality": be- 
havior which is not rational could only be a product of ignorance 
or error. But if the theory is committed to the notion that man's 
only significant mode of cognitive orientation to the situation is 
through scientific knowledge, how then is such irrational behavior 
to be explained? Ultimately, given their premises, only by abandon- 
ing the subjective, action frame of reference altogether and look- 
ing for nonsubjective conditions of action. And these, in turn, 
finally boil down to the traditional notions of heredity and environ- 
ment as conceptualized by the biologists, or to the stimulus-re- 
sponse notions of the behaviorists. In either case, the end result 
is not a general theory of action but some form of positivism. 
Pareto's distinction between nonrational and irrational behavior, and 
his analysis of nonrational behavior as motivated and intelligible, 
seemed to Parsons a giant step toward the solution of this problem. 

Finally, Parsons saw a rather similar dilemma in the way the 
utilitarians and economists dealt with the problem of the goals of 
action. The orthodox or pure theory position was clear enough: let 
them be random and unexplained; take them simply as given data. 
But although this works well enough for various special types of 
analytical theory, it will clearly not do for a general theory of ac- 
tion. And here, once again, many theorists were not willing to let 


well enough alone, but pressed on in an effort somehow to "ex- 
plain" goals, or wants, or ends, and somehow to incorporate them 
into their system. But in doing so. Parsons argued, they usually got 
into serious trouble. Either they attempted to explain goals ration- 
ally, in which case they tended to lose their ideal, normative char- 
acter and became simply predictions of future states of affairs; or 
somehow goals became reduced to parcels of instincts or drives, in 
which case action became explainable without reference to the sub- 
jective process or orientation. In either case, the theory thus reduces 
to one form or other of radical positivism, the former to what Par- 
sons labeled radical rationaUstic positivism, the latter to radical 
anti-intellectual positivism. 


By this point, perhaps. Parsons' grounds for rejecting any form 
of radical positivism should be apparent. All forms of radical 
positivism tend to view the world as a closed, determinate system, 
in principle "explainable" through a rigorous scientific analysis of 
the intermeshings of cause and effect. Parsons would point to social 
Darwinism and also to modern behaviorism as examples of this 
point of view. Make it as sophisticated as you like, as in recent ex- 
tensions of behavior and learning theory; endow the responding 
organism with all kinds of capacities to learn, to acquire new drives 
or to modify old ones, and the system still remains radically 

Push any of these systems to their limit. Parsons was inclined to 
argue, and they cease to be theories of "action" at all. For they 
leave no room for such notions as mind, consciousness, values, ends, 
or normative standards. Action is meaningful, he reasoned, only if 
preceded by a functionally relevant process of orientation, and this 
is possible only if some freedom exists to choose among alternative 
courses. If you are to have any theory of action at all, it must thus 
necessarily make room for an element of voluntarism. Without this 
crucial element of freedom, denied in any closed, determinate sys- 
tem, action becomes mere behavior; subjective and normative fac- 
tors become mere epiphenomena of no causal or explanatory signif- 
icance, and all notions of morahty and responsibihty become mere 

14 • Edward C. Devereux, Jr. 

On yet another ground Parsons took issue with the radical posi- 
tivists. He argued that this sort of theory tended to lead to infinite 
reductionism. In this view, groups are quickly resolved into their 
component individuals, individual personalities are resolved into 
complex organizations of dendrites, neurones, and synaptic mate- 
rials, and these, in turn, to physico-chemical reactions. Say, if you 
like, that the man is going to church; but don't forget that this is 
really only a very complex chain of physical reactions. Whereas 
higher level concepts may be used for descriptive shorthand, ex- 
planations must always be pushed to lower and lower levels. The 
whole is never more than the sum of its parts. 

Parsons could argue on theoretical grounds that this sort of radical 
positivism would not only preclude a science of social action; it 
would also reduce all forms of sociology, and psychology as well, 
to literary exercises of no explanatory relevance. Just as it follows 
that if you are to have any theory of action at all you must accept 
a postulate of voluntarism, so it also follows that if you are to have 
a causally relevant theory of sociology, you must accept a postulate 
of emergence. Parsons clearly accepted both postulates and made 
them central to his own theory. By emergence, he simply means 
that systems have properties which are not reducible or explainable 
in terms of the parts which make them up, and that at various 
levels of organizational complexity ever new orders of systems tend 
to emerge. 

Perhaps we had better follow through a typical Parsonian ex- 
ample. If you are going to talk about action, the smallest meaning- 
ful unit is the unit act, complete with actor, situation, goals, norma- 
tive standards, processes of orientation and choice, and finally, the 
action itself, in which eflPort is expended and means are utilized to 
overcome obstacles and approach the goal. It is thus with the man 
going to church. To be sure, in getting there he expends biological 
energy, moves first one foot and then another, breathes and blinks; 
but, Parsons insists, you will never get to the heart of the matter- 
that is, why he is going to church instead of somewhere else— by 
looking at these alleged "parts." For a science of action, unit acts 
in their totality must be taken as irreducible units. 

But now we may come quickly to the Parsonian notion of the 
emergence of new properties at higher levels by looking at his con- 


ception of action systems. In this connection, consider what is im- 
phed in the notion of "economic rationaHty." Whereas a single act 
may be technically rational with respect to its own single goal, 
there is no meaningful way in which the economic rationality of a 
unit act may be determined without reference to a system consist- 
ing of two or more goals of the same actor. For the minimum and 
irreducible notion involved in the concept of "economizing" is that 
of counting the costs of resources allocated to any one goal in terms 
of their alternative uses for other goals. Hence, unless there are at 
least two goals or values, and unless these are articulated, however 
crudely, into some single system, and unless there is an actor who 
has both these goals as his own, and hence who must be presumed 
to persist at least a httle while through time with at least a modicum 
of identity— unless all these conditions exist, then it is simply mean- 
ingless to talk about economic rationality at all. For economic 
rationality appears as an irreducible property of action systems, 
and applies to unit acts only as components of some such larger 

Now, if these theoretical grounds for Parsons' election to talk 
about the emergent properties of action systems are clear enough, 
consider the empirical grounds. Very simply. Parsons argues, eco- 
nomic rationality may be empirically demonstrated to function as 
at least one of the major determinants of economic behavior. If 
this were not the case, then none of the deductively elaborated pre- 
dictions of economic theory would have any empirical relevance 
whatever; but they do. In spite of the many faults of economic 
theorizing and in spite of the disturbances caused by the operation 
of numerous nonrational factors, it is simply a fact that concrete 
market systems do reflect in rough approximation the operation of 
economic laws premised upon the notion of individuals and firms 
acting with economic rationality. 

In his later work. Parsons has used essentially this same form of 
argument in demonstrating the emergent reality and causal rele- 
vance of many other system properties, at the levels of personality 
organization, social structure, and culture. In each instance, the 
argument turns upon the recognition of properties which are not 
analytically reducible to their component parts or elements, or 
which behave in ways which cannot be inferred from a study of 

16 • Edward C. Devereux, Jr. 

the parts outside the system, and then of attempting to estabUsh 
the causal relevance of the phenomena thus identified. When this 
is done, he reasons, it is no longer possible to regard such properties 
as mere descriptive tags or as epiphenomena. 

We have reviewed Parsons' grounds for rejecting any radically 
positivistic position as an adequate or sufficient basis for the de- 
velopment of sociological theory. Yet Parsons did not argue that, 
therefore, everything which the positivists had done was either 
wrong or irrelevant. Although he was impressed with the principle 
of emergence and with the element of indeterminacy implied in the 
voluntaristic postulate, he was also impressed with the fact that 
emergent systems never wholly detach themselves from their more 
primitive parts or elements. Even the best socialized human being 
is still, among other things, a concrete physiological organism, and 
presumably the new-born baby is only that. The stubborn facts of 
heredity and environment are always there, as crucial parameters 
for the human personality and the social system alike, and their 
particular forms always use up many important degrees of freedom. 
Emergent systems are thus never wholly free-floating. The prob- 
lem, as Parsons saw it, was to construct a single theoretical system 
which could handle both types of factors and work out in detail 
the points of articulation and interaction between them. He saw 
the voluntaristic theory of action as enabling the theorist to do 
just that. 


In view of the limitations which Parsons saw in any radically 
positivistic approach to social behavior, we might expect that he 
would have responded more favorably toward the conceptions of 
human behavior formulated by the German idealists. Although 
there are of course important diflFerences in the specific theories of 
Kant and Hegel, or of Dilthey, Spengler, Sombart, or Tonnies, all 
of these writers tended to share in common some notion that hu- 
man action could not be adequately explained by the interactions 
of causal laws of the sort which presumably determined the flow of 
events in the world of natural phenomena. All were concerned with 
the uniquely "human" quality of social action, and seemed to agree 
that this human quality somehow revolved about the elements of 

parsons' sociological theory • 17 

freedom, spirit, values, or morality. They were inclined to regard 
social structure not as a factual or causal order, but as a moral order, 
or as a system somehow expressing and actuaUzing certain key 
values embedded in the Geist or spirit of the people. Relationships 
of parts within the whole were regarded as being governed not so 
much by laws of causal interdependence as by norms of logico- 
meaningful coherence. 

Parsons did indeed find much of value in the formulations of the 
idealists. Clearly they were attempting to deal with important 
phenomena which were dismissed as simply irrelevant by the 
positivists and which had never been adequately conceptualized 
by the utilitarians. He was convinced that any adequate general 
theory of social action would have to find some way of taking ideas, 
ideals, norms, values, and ends into account, not simply as given 
facts, but explicitly as ideal elements. Perhaps at lower levels of 
behavior organization, for example in accounting for the movement 
of moths toward a light or the reflexive blinking of the human eye, 
mechanistic stimulus-response explanations might do. But at the 
level of human action in organized social systems, in Parsons' view, 
such explanations would never tell the whole story. At this emergent 
level, he was convinced, empirical evidence conclusively demon- 
strates an important causal role for ideas and ideals, and hence any 
general theory must be able to handle them as independent 

Yet Parsons also saw very serious difficulties in the way these 
elements had been treated by the idealists. It seemed to him that 
most theorists in this tradition, from Hegel right on through Ruth 
Benedict, placed too strong a stress on the role of ideal elements, 
at the expense of other relevant considerations. It was perhaps too 
simple: find the spirit or Geist of a particular culture, show how 
certain patterns or themes run through it in logico-meaningful con- 
figurations, and all, or nearly all, is thereby explained. 

If each society must be explained and interpreted in terms of its 
own unique Geist or spirit, then there could be no general theory 
or general laws which could apply across societies. Social science, 
in this case, finds itself reduced to a kind of historicism, in which 
all efforts are expended in the exhaustive description and interpreta- 
tion of unique historical situations. And yet the whole point of a 

18 • Edward C. Devereux, Jr. 

social theory, for Parsons, is that it should be a general, analytical 
theory, permitting systematic comparisons of all societies and the 
development of general laws about them. 

The tendency of the idealists to maneuver their ideal elements 
into a position of simple and sovereign primacy has still another un- 
fortunate consequence. If all concrete social phenomena are re- 
garded as direct emanations or expressions of the given value con- 
figuration, the analysis is likely to sidestep the difficult and crucially 
important task of working out in detail the mechanisms through 
which these ideal elements become articulated in concrete be- 
havior. In particular, one is likely to miss the dynamic significance 
of situations in which there is not a neat, one-to-one fit between 
value elements and on-going social systems. And yet it is precisely 
these situations, as for example in the rise of new prophetic re- 
ligious movements, which offer the most fruitful leads for an analy- 
sis of the relationships between ideas and social structure. 

It also seemed to Parsons that in the idealistic theories, ideas 
were treated as altogether too free, and hence essentially as free- 
floating. Analysis of the important questions concerning where they 
come from, how they develop and change, how they are learned 
and transmitted, how they interact with each other and with the 
more stubborn factual levels of social structure and of heredity 
and environment thus tends to be by-passed. Parsons was too well 
versed in Marx to accept the Hegelian notion that cultural ideas 
constituted wholly self-contained systems, developing only by the 
dialectical unfoldings of their own inner logic or given direction by 
some vague notions concerning an over-riding Weltgeist. And yet 
he was too well versed in Pareto to dismiss such forces as the 
pressure toward ethical or logical consistency as simply irrelevant. 
Max Weber's solution, in which the relationship of ideas and on- 
going social structure was handled as one of mutual interdepend- 
ence, seemed to Parsons a step in the right direction. 

I have attempted to review what Parsons saw as assets and 
liabilities in each of the three schools of thought with which he 
concerned himself during his first decade of critical scholarship. 
With respect to the utilitarians and orthodox economists, he ad- 


mired their analytical elegance and their action frame of reference, 
but felt that their strong individuaHstic, rationalistic biases stood 
in the way of the development of a general theory of action ade- 
quate to handle the problems of order, of nonrational action, of 
goals or wants, and of normative standards. With respect to the 
positivists, he saw as fruitful their attempts to deal with the physi- 
cal and physiological parameters of personality and human be- 
havior, but rejected the elements of mechanistic determinism and 
reductionism implicit in any radical generalization of this approach. 
With respect to the idealists, he welcomed their analysis of cultural 
configurations and of the role of ideas, values, and norms, but 
argued that their treatment of these elements was too one-sided 
and free-wheeling, and that the postulate of cultural relativity led 
to a kind of historicism which blocked the development of general 

What Parsons saw as necessary was a single general theory that 
would incorporate the permanently valid precipitate in each of 
these approaches while at the same time overcoming the limita- 
tions implicit in each. He believed that there were clear signs of a 
convergent movement toward such a theory in the works of cer- 
tain recent representatives of each of these approaches. In each 
case, the end point of convergence seemed to be upon what Par- 
sons elected to call a voluntaristic theory of action. Virtually all of 
his subsequent work has been devoted to the systematic develop- 
ment and elaboration of this theory. 

Let me attempt to summarize the basic orientations which Par- 
sons seems to have derived from his critical studies. (1) The ob- 
jective is always to construct an adequate general theon,^: for 
Parsons this means a theory which is elegant, analytical, systematic, 
and complete in the sense that some place is found in the theory 
for all of the types of factors concretely relevant to the operation 
of the empirical system, including those which are treated as param- 
eters or simply ignored in various types of special theories. (2) 
An adequate general sociological theory must be an action theory: 
for Parsons this means that the central mechanism must always be 
some notion of actors orienting themselves to situations, with refer- 
ence to various sorts of goals, values, and normative standards, and 
behaving accordingly. (3) Any meaningful action theory must be 

20 • Edward C. Devereux, Jr. 

based on a voluntaristic postulate: for Parsons this means simply 
that choice among alternative values and courses of action must 
remain at least partially free. By implication, it seems to follow 
that human action systems can never become empirically closed. 
(4) In a voluntaristic theory of action, ideas, ideals, goals, and 
normative standards must be treated as causally relevant variables, 
and not as epiphenomena. These ideal elements are regarded as 
mutually interdependent empirically with the various nonideal 
elements in the empirical world, but this interdependence never 
uses up all of the degrees of freedom: the content of the ideal 
variables is never wholly determined by the pressures of nonideal 
forces and constraints. (5) Sociological theory must take into ac- 
count the principle of emergence: for Parsons this means that at 
various levels of organizational complexity, systems emerge which 
have properties which cannot be inferred from or explained in 
terms of the operation of their component parts or elements, and 
that these emergent properties must be treated as causally relevant 
variables in the theory. By implication, at each emergent level cer- 
tain new degrees of freedom are created. (6) Systems and their 
emergent properties never become wholly detached from their own 
component parts; important areas of mutual interaction and em- 
pirical interdependency function to limit degrees of freedom on 
both sides. For Parsons, this clearly implies not only that emergent 
properties are limited, though not deteiTnined, by the nature of 
the system from which they emerge, but also that the nature of the 
parts may themselves be significantly altered, though not without 
limit, by the operation of the emergent variables. 

These, to this reviewer, seem to be the basic orientations which 
Parsons carried forward from his critical review of the literature 
and which have served as guiding principles in his own efforts at 
theory construction. 


Parsonian theory, as we have observed, is based upon an ac 
tion frame of reference. Where others talk of organismi and environ 
ment. Parsons talks of actor and situation. Where others talk of be 


havior or response, Parsons talks o£ action. All action, to be sure, is 
behavior; but not all behavior is action. If the flight of a moth 
toward a candle is conceived simply as a mechanistic response of its 
organism to the stimulus of light, there is behavior but not action. 
On the other hand, if we were to conceive of some subjective proc- 
ess of orientation as an essential link in the chain ... as if the moth, 
for example, were to reason with itself: "What a pretty hght! I 
would like to be closer to it. I will fly there as directly and quickly 
as possible . . . ," then we should be deaHng with action. Typically, 
of course, we do not feel it necessary to make such imputations 
with respect to most of the behavior of lower organisms, nor indeed 
to certain areas of human behavior. Parsonian theory is simply not 
interested in such behavior. His entire theory rests upon the prem- 
ise that there are broad areas of human conduct which do in fact 
properly qualify as action, and that these are the kinds of be- 
haviors which most legitimately concern the sociologist. The mini- 
mum frame of reference for talking about action must therefore 
include, besides the actor and the situation, some explicit reference 
to subjective processes or orientation, conceived as causally relevant 
intervening mechanisms and not as epiphenomena, and to ex- 
plicitly formulated notions of ends or goals and of normative 
standards, conceived as ideal elements which function to structure 
the actor's orientation to situations. 

In Parsonian theory, the actor is taken as an analytical point of 
reference somewhat akin to the ego of Freudian psychology: it is 
the executive officer which perceives, evaluates, and organizes ex- 
perience and controls the approaches to motility. While concrete 
actors typically dwell in concrete organisms, it is essential to Par- 
sonian analysis that the actor's own body, with its various needs 
and capacities, may be defined by the actor as a part of its situa- 
tion. The relationship of organic needs to the goals of action is left 
open to empirical investigation. 

The situation, in action theory, is not simply the sum of sensory 
stimuli impinging upon the actor at a given moment, but rather 
something which is both more and less than this. Essentially it con- 
sists of whatever is meaningfully organized in the actor's orienta- 
tion. In defining his situation, the actor may take into account cer- 
tain objects in his immediate surroundings— whether physical, 

22 • Edward C. Devereux, Jr. 

social, cultural, or symbolic— while dismissing others as irrelevant. 
But he may also include in his situation a variety of objects not 
physically present at the moment— for example, absent persons or 
predicted future events. It follows that the Parsonian situation 
cannot be defined independently of the actor's orientation toward 
it. It is not, however, a purely free-floating figment of the actor's 
imagination: for his orientations clearly make reference to objects 
outside himself and beyond his control. He may, of course, misde- 
fine the situation; but if he goes too far wrong, reality will pre- 
sumably find ways to strike back at him. He is thus under some 
pressure to keep his orientations in -some adjustment to the facts 
of the external world. 

As we have seen, the subjective process of orientation plays a 
central role in action theory. In this complex process of defining 
the situation, as Parsons has analyzed it, the actor constructs a 
cognitive map of the situation and appraises or evaluates it in 
terms of its relevance to his various goals, interests, and normative 
standards. In this process, the situation is structured by the actor 
into some meaningful configuration, in which its various elements 
are seen as things to be desired or to be avoided, as obstacles to be 
overcome, as conditions to be accepted, or as potential means to be 
utilized. The actor must predict how the situation may be expected 
to develop and consider whether, in terms of his own goals and 
values, some active intervention is necessary or feasible; he must 
consider alternative courses of possible action, and predict and 
evaluate their consequences in terms of his various goals and 
normative standards. 

Although this process is most explicit in the special case of 
rational action directed toward an empirical end. Parsons maintains 
that the same schema is also useful for analyzing various sorts of 
nonrational action as well— for example, action directed toward 
some nonempirical goal such as salvation, or activity which is 
primarily expressive of some vague value-attitude. Even though 
many steps in the full process of orientation may be short-circuited, 
even though the resultant orientation may be distorted by the press 
of unconscious factors or badly out-of-line with reality, if there is 
still some recognizable "definition of the situation' and if action 


occurs as if premised upon it, the action frame of reference is held 
to be appropriate for its analysis. 

Goals or ends, in the Parsonian schema, are simply the actor's 
pictures of future states of affairs regarded as desirable and worth 
striving toward, or in the negative case, as undesirable and worth 
guarding against. They may be immediate or remote, highly specific 
or vague and general, objective, subjective, or even transcendental. 
Ice cream sodas, college degrees, success, happiness, or salvation 
may all be treated as goals to the extent that the actor has cathexis 
toward them. 

Typically, of course, the actor has many different goals, and in 
order to allocate time and resources among them he will have to 
have various standards for evaluating their relative importance: 
work before pleasure, or poetry before pickles. And typically, more- 
over, any given course of action will have implications for many 
different goals or values of the same actor, a fact which requires 
the actor to apply a variety of normative standards to his contem- 
plated plan of action. In appraising an action in terms of the norms 
of technical effectiveness, of economic efficiency, of moral worth, 
or of aesthetic appropriateness, the actor in effect is bringing it into 
alignment with his total system of values. 

In principle, one might employ the action schema for the analysis 
of unit acts, but Parsons is not much interested in this. For him, 
the important point is that practically all action occurs in systems. 
If we argue that actors typically have many different needs, goals, 
and values, and that unit acts typically have consequences for many 
of these, we are involved at once with the systemic nature of ac- 
tion. The issue of pickles versus poetry is meaningful only if these 
be thought of as values of the same actor, who has some identity 
through time and who thus must struggle to put each in proper 
perspective in relation to some total system of values. Situations 
also typically persist in time or have a recurrent character, so that 
the actor may carry forward orientations learned in the past to 
help define present situations. As relationships between the actor 
and certain recurrent aspects of his situation thus become stabihzed, 
action itself develops a recurrent character and we are deahng 
with an action system. In time, moreover, the various action sys- 
tems of a particular actor become themselves more or less organized 

24 • Edward C. Devereux, Jr. 

into a single, more complex system, constituting in effect what 
W. I. Thomas would call the individual's life organization. As we 
shall see, this total action system of a particular actor occupies a 
central position in the Parsonian conception of personality. To 
visualize such a system in its simplest form, we might picture the 
patterned and recurrent sets of activities carried out by Robinson 
Crusoe after he had finally worked out a way of coming to terms 
with his island environment, but before Friday had made his ap- 

In principle, one might take the goals and normative standards 
employed by a particular actor simply as given data. In a general 
theory of action, however, one must ask where they come from and 
what determines their particular content. In keeping with his 
voluntaristic postulate, Parsons argues that they are at least in part 
free creations of the individual actor: without this degree of free- 
dom, they would lose their postulated ideal character and would 
have no relevance as independent variables. It does not follow, 
however, that the actor is free to fashion his goals and values out 
of whole cloth. For in shaping them, Parsons argues, he is under 
some constraint to deal, on the one hand, with his own biological 
and psychological needs, and on the other, with the normative 
systems of his sociocultural environment. 

With respect to the former. Parsons is much concerned with the 
ways in which various needs of the organism become articulated 
with appropriate patterns of activity which bring them into rela- 
tionships with goal objects suitable for their satisfaction. Thus the 
hunger of the infant develops into the goal of food seeking and is 
quickly embedded into an action system which must take into ac- 
count other values and conditions as well, such as table manners or 
relations with mother. Where a need has become linked to a suita- 
ble goal-object and with a disposition to carry out certain patterned 
activities with respect to it. Parsons refers to emergent need-dis- 
positions. It is central to Parsons' thinking that these need-disposi- 
tions, which function to bring needs into focus as situation-oriented 
goals, involve an element of learning and that they are modifiable. 
Moreover, while there may in fact be some irreducible list of or- 
ganically given needs, it appears that many others are acquired as 
derivatives of social experience— for example, the needs for security 

parsons' sociological theory • 25 

or affection. Although the actor is under strong pressures to engage 
in activities appropriate to his own needs, in setting his goal-sys- 
tem he still has some freedom to push them around a bit: the 
martyr may choose to starve himself, the celibate to forego sex. 

With respect to the second source of constraint. Parsons argues 
that individual actors are under some pressure to develop goals 
and standards which are in accord with those of the sociocultural 
system in which they live. Although Robinson Crusoe may have had 
a relatively free hand in developing his own life-organization, 
clearly the child born into an on-going society does not. The situa- 
tion of the child is stablized and organized for him by his parents, 
and he is continually presented with ready-made definitions which 
he is under some constraint to accept as his own. Ideally, the end- 
product of socialization is an individual who has successfully in- 
ternalized the culture goals and normative standards of his society 
and who has worked out a pattern of activities which serves in- 
dividual and societal values simultaneously. It is central to Parsons' 
thinking, however, that this process of socialization is never really 
complete, and that what is achieved is not achieved without cost: 
there always remains some lack of congruence between individual 
and societal goals. Though obviously influenced both by psycho- 
logical and cultural pressures, individual goals are not wholly de- 
termined by either or by both together. 

In principle, the action schema we have described could apply 
to a solitary actor on a desert island. In fact, most human action, 
and especially that which is of interest to sociologists, occurs in 
society and has other persons as significant objects in the situation 
of the actor. In this case, the action frame of reference broadens to 
become one of interaction. 

The simplest case is that of the dyad: ego has alter as a signif- 
icant object in his situation and alter has ego as an object in his 
own. In acting with respect to alter, ego must predict how he will 
respond; in effect, his action is designed to produce a certain de- 
sired reaction in alter. And of course alter is presumably doing the 
same sort of thing with respect to ego. Interaction thus has the char- 
acteristic which Parsons has called double contingency. If the two 
are not well acquainted, we may expect that there will be many 
wrong predictions at first, and many communication failures. But 

26 • Edward C. Devereux, Jr. 

after a while, Parsons observes, they may get to be rather good at 
it. Their actions with respect to one another tend to become pat- 
terned and stabilized: when interacting, ego comes to play a specific 
sort of role in relation to alter and expects alter to play a specific 
sort of role in relation to himself. In some such manner, a child and 
his mother learn what to expect of one another and we have a 
miniature two-role social system, in which each role is com- 
plementary to the other. 

Presumably, if the relationship endures, each member of the sys- 
tem derives certain satisfactions and meets certain needs through 
his participation in it: to this extent, each develops a vested inter- 
est in the continuity and stability of the relationship. Where this is 
true, alter's conformity to role expectations will come to have re- 
ward value for ego, and his deviations will quickly be countered 
with negative sanctions of some sort. And alter, of course, would do 
the same for ego. The eflFect of this mutual sanctioning is to create 
a mechanism which operates to preserve the equilibrium of the 
miniature social system: minor disturbances set forces in motion 
which function to restore the status quo ante. In the course of time, 
moreover, the dyad is likely to develop its own private culture, 
consisting of shared bits of knowledge, techniques, symbols with 
special shared meanings, tools and other significant objects, norma- 
tive standards and even goals. Culture, in this sense, thus represents 
the shared property of the members of the social system: the items 
which comprise it are all potentially teachable or transferable to 
some new member of the system. 

In some such manner as this, Parsonian theory moves from an 
action frame of reference to one of interaction and thence to the 
concepts of social system and culture. The primitive, spontaneously 
developing social system we have pictured contains all the prop- 
erties which Parsons holds to apply more generally to any social 
system: two or more actors occupying differentiated statuses or 
positions and performing differentiated roles, some organized pat- 
tern governing the relationships of the members and describing 
their rights and obligations with respect to one another, and some 
set of common norms or values, together with various types of 
shared cultural objects and symbols. Parsons postulates of social 
systems that they are boundary -maintaining, in the sense that there 

parsons' sociological theory • 27 

tends to be a tighter, more integrated organization among the com- 
ponents of the system, while it is operating as such, than there is 
between these components and elements outside the system. And, 
as we have seen, he also postulates an equilibrium tendency: in- 
deed, unless the system has built-in mechanisms which function to 
hold it in some sort of steady state over a period of time, it is hardly 
worth designating as a system at all. The nature and extent of 
equilibrium in any particular system, however, are left open for 
empirical investigation. The defining properties of social systems 
are thus conceived as differentiation, organization, boundary main- 
tenance, and equilibrium tendency. 

We have indicated how Parsons conceives social systems to de- 
velop spontaneously whenever two or more actors come into some 
stabilized, patterned mode of interaction. It was presumably thus 
with Crusoe and Friday; the process may also be observed in the 
sorts of ad hoc experimental groups analyzed by R. F. Bales and 
his associates.* In on-going societies, however, virtually all inter- 
active systems develop within the matrix of an already established 
sociocultural system which is not altogether silent about the defini- 
tions of roles, normative standards, and system-goals. When two 
people marry, they form a new family, but they do not thereby 
invent the institution of the family. Both partners bring to the new 
relationship a host of notions about the roles which each should 
play, and may punish behavior perceived as deviant from these 
expectations. The couple, moreover, experiences sanctioning pres- 
sure from the surrounding community to bring it into some con- 
formity with societal expectations. 

In the end, of course, each couple has to work out its own final 
set of mutual adjustments, in which the socially prescribed roles 
are modified and embellished in various ways. In effect, every con- 
crete role relationship thus involves an institutional nexus and a 
particular nexus, and it is useful for analytical purposes to keep these 
two components separate. Parsons uses the term social role to refer 
to the institutionally defined and regulated component of roles. Be- 

* See R. F. Bales, "The Equilibrium Problem in Small Groups," and 
Philip E. Slater, "Role Differentiation in Small Groups," reprinted in 
A. Paul Hare, et al., eds., Small Groups: Studies in Social Interaction 
(New York: Alfred A. Knopf, Inc., 1955), pp. 424-456 and 498-515. 


28 • Edward C. Devereux, Jr. 

cause o£ their institutionalized character, it follows that social roles 
may be analyzed independently of specific knowledge about any 
particular incumbents; they tend, moreover, to have a normative 
character and to be generalized across many concrete social sys- 
tems within a given society. When Parsons talks about the middle- 
class family or the doctor-patient relationship, he is usually deal- 
ing with these institutionalized components of roles. 

In yet another way concrete social systems are seen as embedded 
in a larger social matrix: small systems are usually components of 
larger ones, which in turn are components of still larger ones. Thus 
the mother-child dyad is a sub-system within the family, and the 
shipping room a sub-system within the factory. When he is dealing 
with more complex systems. Parsons treats as components not in- 
dividuals but various sub-systems which are now regarded, ana- 
lytically, as the actors. Thus in the sort of input-output analysis he 
has been pursuing in recent publications, each department in a 
business firm would be treated as if it were a single actor, perform- 
ing certain roles in relation to other departments, receiving certain 
inputs and delivering certain outputs. At this level of analysis. Par- 
sons reasons, what goes on within a particular sub-system need not 
concern us directly: only the product matters. Incidentally, he fol- 
lows the same principle when dealing with social systems in which 
the actors are concrete persons: when focusing upon the social 
system, talk about roles and relationships, not about processes in- 
ternal to the personalities involved. 

Finally, we should observe that, just as the roles in a newly 
formed social system tend to be drawn from and molded by the 
institutional system of the surrounding society, so also the culture 
of the social system tends to draw upon or incorporate various ele- 
ments in the broader culture surrounding it. Perhaps Crusoe and 
Friday had a relatively free hand in this respect, but it was probably 
also rather tough going to hammer out from scratch an appropriate 
set of shared symbols, values, meanings, and so on. The existence 
within an on-going society of a large repertory of standardized 
cultural objects and symbols probably functions not only to facili- 
tate the development of social systems but also to limit variation 
among them. 

parsons' sociological theory • 29 


I have attempted to sketch in rough first approximation how 
Parsonian theory builds upon an action and interaction frame of 
reference to generate certain basic notions regarding social sys- 
tems, personality, and culture. The reader will have observed that 
it is difiicult to say very much about any one of these concepts 
without becoming involved with the others as well. The reason for 
this is that, in Parsons' view, these concepts refer to systems which, 
while analytically separable, nevertheless empirically interpene- 
trate one another. These different orders of analytical systems are 
regarded as jointly participating in and partly determining process 
in the same concrete empirical action systems. 

How Parsons conceives of the phenomenon of interpenetration 
may be illustrated in the way he elects to handle the ancient body- 
mind problem.* Concrete human personalities always reside in con- 
crete organisms, but to Parsons this does not imply that personality 
organization and process are therefore somehow reducible to physio- 
logical structure and process, and that the laws worked out at this 
organic level need only be applied to the next higher level to arrive 
at a complete explanation. It merely means that, because the t^^'0 
orders of systems are empirically interpenetrating, there must be 
identifiable physiological mechanisms for all the processes operative 
at the psychological level. 

Why then must we deal with two analytical systems instead of 
one concrete one? Partly, Parsons seems to argue, as a matter of 
analytical convenience and efficiency: you can't talk about ever\^- 
thing at once, and you will make more progress by focusing on one 
aspect at a time. His more fundamental answer is: each order of 
system also represents in part an emergent empirical system with 
its own unique organization, characterized by a selective inclusion 
of elements drawn from lower-order systems and by a distinctive 

* "An Approach to Psychological Theory in Terms of the Theory of 
Action," in Sigmund Koch, ed., Psychology: A Study of a Science 
(New York: McGraw-HiU, 1959), Vol. 3, pp. 647-651 

30 • Edward C. Devereux, Jr. 

pattern which it imposes on the relationships of the selected com- 
ponents. Thus the personality system, for Parsons, is distinctly a 
psychological and not a biological system; at its own level, it has 
its own system problems— relative to such functional exigencies as 
drive reduction, maintenance of repressions, integration, and so on 
—its own boundaries and boundary maintenance problems, and its 
own equilibrium tendencies. 

By conceptualizing organism and personality as two analytically 
separate orders of systems, partially interpenetrating but partially 
also independent. Parsons seeks to gain a point of leverage for 
analyzing the empirical relationships between them. This is how 
he conceives the outputs of each system with respect to the other. 
The personality is pictured as receiving, as outputs from the organic 
system, such facilities as motivational energy, perceptual capacity, 
performance or response capacity, and a sort of integrative facility 
rooted in the mechanisms of learning. In return, the personality 
generates outputs with respect to the organic system in the form of 
"motive force," conceived as a sort of feedback process in which 
motivational energy contributed by the organic system is controlled 
by the psychological system and brought to bear upon instrumental 
processes; performance potentials are thereby greatly increased. 
More specifically, Parsons conceives of the personality system as 
contributing a directional component and an attitudinal set which 
function to focus perception and guide goal-seeking activities. The 
attitudinal set creates an expectation that, as psychological needs 
are met, organic interests will also be served. In this connection, 
one might think of the complex processes implied in the pattern of 
deferred gratification involved in long-range goal-seeking activity. 

Parsons conceives of the interrelationships of personality sys- 
tems, social systems, and cultural systems along essentially similar 
lines. Each is regarded as an analytically separate order of systems, 
partially independent, but partially interpenetrating the others, in 
such a way that all three participate jointly in the determination 
of concrete action systems. Although for certain analytical purposes 
we may focus on one order of system at a time, a general theory of 
action will necessarily involve systematic reference to all three. 

Consider, in this connection, the Parsonian conception of culture. 
To Parsons, it makes sense to think of cultural phenomena as form- 


ing systems in their own right, with their own laws of internal 
organization and development. Thus the cultural scientist may 
legitimately devote his attention to the study of linguistic systems, 
or of ethical or rehgious systems, or of philosophical, scientific, or 
legal systems, without becoming much involved with sociological 
or social-psychological considerations. Many of the laws of lin- 
guistic development, for example, hold almost wholly independ- 
ently of cultural or sociological contexts. Ethical systems move as 
if driven by a strain toward consistency. Therefore, Parsons rea- 
sons, cultural systems have emergent system properties of their 
own, and enjoy at least some measure of autonomy in their develop- 

But this does not mean, for Parsons, that cultural systems ever 
become wholly detached or free-floating, as in the idealist view. 
Lock them up in libraries or museums and nothing much happens. 
In his most recent discussions of this problem. Parsons has moved 
from a view of cultural systems as relatively detached object-sym- 
bol-meaning configurations, toward a view which holds them to be 
special sorts of action systems, organized about the specific func- 
tional exigency of maintaining symbol-meaning systems.* 

Clearly, Parsons argues, there is ample evidence of mutual 
empirical interdependencies between culture on the one hand and 
personality and social systems on the other. Much of the content 
and direction of movement in cultural systems reflects functional 
problems arising at the level of these lower order systems: con- 
sider, for example, the projective character of some cultural sys- 
tems, as evidenced in the works of Kardiner and Whiting. Marx's 
analysis of "ideologies" and Pareto's work on "derivations" also point 
to ways in which cultural systems reflect problems of social structure. 
As evidence that culture also reacts back upon the social system. 
Parsons would point to Max Weber's analysis of the consequences of 
the Protestant ethic for the development of capitalism. 

With respect to interpenetration, Parsons argues that all on-going 

* In this connection see A. L. Kroeber and T. Parsons, "The Con- 
cepts of Culture and Social System," American Sociological Review, 
XXIII (October 1958), pp. 582-583, and Parsons' reply to comments 
of R. H. Ogles and M. J. Levy on this paper, in "Culture and Social 
Systems: An Exchange," ibid., XXJV (April 1959), pp. 248-250. 

32 • Edward C. Devereux, Jr. 

social systems must have a culture, in the sense of shared norms, 
meanings, symbols, and objects, and that these are usually drawn 
from and related to those in the parent culture. Still more impres- 
sive, in Parsons' view, is in the interpenetrating relationship of 
institutionalization at the cultural level and internalization at the 
personality level. Whereas it may appear to a particular unaccul- 
turated individual that cultural norms are simply external and 
constraining aspects of the situation, Parsons argues that this could 
hardly be generahzed for all members of a society. Taking a lead 
from Durkheim, he reasons that the moral force of institutionalized 
norms depends, in the final analysis, upon the internalized con- 
science of a core of culture bearers who mobilize in their behalf 
the powerful mechanisms of the guilty conscience and righteous 

But let us return for the moment to the level of personality. We 
have already indicated how Parsons pictures the personality system 
as an emergent level of psychological organization, standing in a 
relationship of empirical interdependency and also of interpene- 
tration with the biological organism in which it is housed, but yet 
enjoying also an element of autonomy and having to face system 
problems at its own level. On the other side. Parsons views the 
personality as equally involved in relationships of interdependency 
and interpenetration with social and cultural systems. Needs are 
conceived as developing into need-dispositions and brought into 
focus as goals; and these in turn come into relationship with sys- 
tems of activity carried out in recurrent, meaningfully structured 
situations. Virtually all of these, Parsons observes, are social in 
character and incorporate culturally standardized definitions. In 
the theory of action, as noted before, actor and situation are always 
taken together as forming a single reference system. It follows that 
personality, which Parsons defines as the total action system of a 
single living organism, incorporates in itself a complex set of object- 
relations. Since most of these are sociocultural in character, Parsons 
finds it practically impossible to conceive of the adult human per- 
sonality in other than social terms. 

But even though the stamp of society is everywhere upon it, 
Parsons does not regard the personality simply as a social product, 
as simply the sum and organization of the roles any particular 



individual has learned to play. For in spite of its extensive empirical 
interdependencies and interpenetrations both with sociocultural and 
with biological systems, Parsons maintains that personality con- 
stitutes an important level of system organization in its own right, 
and hence can never be reduced to or fully explained by the other 
systems to which it relates. There remains at least some small area 
of autonomy or freedom, perhaps now in the form of the divine 

Much of the Parsonian argument with respect to social systems 
has already been implied in what has been said before. It is clear 
that the same concrete activities which are carried out by indi- 
vidual personalities appear again as performances in the social 
system frame of reference. Yet it is central to Parsonian theory that 
social systems, although involved in extensive relationships of 
empirical interdependence and interpenetration both with person- 
ality systems and with cultural systems, also represent emergent 
entities with their own system problems, boundaries, and equilib- 
rium tendencies. 


The problems of order, integration, and equilibrium have al- 
ways played a central role in Parsons' thinking. Indeed, some critics 
have charged that an excessive concern with these problems has 
tended to give his theoretical system a static, conservative bias. Par- 
sons does indeed postulate an equilibrium-seeking tendency as a 
property of systems of any sort, partly as a generahzation from experi- 
ence, but more particularly for heuristic purposes. To this reviewer, 
it appears that Parsons' concern with equilibrium does not reflect 
the view that everything is automatically integrated and adjusted 
to everything else in this best of all possible worlds. It reflects 
instead the view that society represents a veritable powder keg 
of conflicting forces, pushing and hauling in all ways at once. That 
any sort of equilibrium is achieved at all, as it evidendy is in most 
societies most of the time, thus represents for Parsons something 
both of miracle and challenge. Far from taking societal equiUbrium 

34 • Edward C. Devereux, ]r. 

for granted, he sees it as a central problem demanding detailed 
analysis and explanation. 

As Parsons views it, society is not a neatly articulated "organic 
system" in full control of its own internal processes and mechanisms. 
It consists instead of a loosely federated congeries of systems and 
sub-systems of many different sorts, each, as we have seen, with 
its own internal system problems and equilibrium tendencies, and 
each with its own crucial degrees of freedom. Yet these are con- 
ceived as standing in relationships of interdependence and inter- 
penetration with one another. The result is that almost any concrete 
pattern of actions has consequences for many different sorts of 
system -referents; but no particular course of activity. Parsons has 
argued, can serve simultaneously and with maximum effectiveness 
all the needs of all the systems upon which it impinges. And needs 
which remain unmet for even a while become sources of strain 
and tension, with potentially disruptive consequences. 

Consider in this connection the complex set of functional prob- 
lems with which any social system must somehow come to terms if 
it is to establish and maintain some equilibrium position. First, the 
social system cannot be radically incompatible with the needs, 
motives, and capacities of the human agents who must play its 
various roles. Some social systems have as their primary goal the 
servicing of the needs of individual participants, but many others 
do not. For example, the U.S. Navy is not organized primarily as 
a device for providing gratifications for the officers and sailors. Yet 
somehow it must cope with its human materials. As biological 
systems, the men must be fed and clothed and protected from 
extremes of fatigue or inclement weather. As psychological systems, 
they must be provided sufficient gratifications in terms of their own 
personal needs and goals to check tendencies toward deviance and 
to motivate adequate role performances. The Navy is not a closed 
society, and hence it must contrive to recruit its personnel from the 
broader society with an eye to required capacities and skills. And 
since these are not found ready-made, it must also provide suitable 
devices for training and socialization. Because it does not possess 
its personnel in any inclusive sense, moreover, it must also come 
to terms with the other role-commitments of its actors, who are 
simultaneously members of families, communities, political parties, 

parsons' sociological theory • 35 

and so on, all with their own and sometimes conflicting claims upon 
time and loyalties. 

Now none of these functional problems, relating to what Parsons 
sometimes calls the "motivational problem of order," will solve 
themselves automatically. One cannot ever take for granted. Par- 
sons argues, that the motives, goals, capacities, and values of 
individual actors will automatically move them toward the sorts 
of adequate role performances necessary for the functioning of 
this or that particular social system. It is more nearly correct to 
assume the opposite— that tendencies toward deviance or alienation 
are somehow endemic, rooted as it were in the sheer cussedness 
and variability of human nature. Human beings are born in society 
but not of it. With this as his working postulate, Parsons is guided 
to analyze in detail not only the sources of deviance and strain but 
more particularly the mechanisms of social control and socialization 
by which a social system manages to hold deviance in check and 
enlist the motivations of its participants. 

A second order of functional problems arises from the relation- 
ships of any particular social system to its parent culture. Parsons 
argues that any particular social system will tend to develop a 
set of normative patterns which are somehow relevant for its own 
particular functions. Those which are established to govern rela- 
tionships in the navy are presumably relevant for the coordination 
of activities in a fighting organization, but they will be of a rather 
different sort from those which are supposed to apply in the family 
or in a business concern. Yet all of these differentiated social sys- 
tems coexist in the same concrete society. For Parsons this situation 
represents another potential threat to the equilibrium position of 
any particular social system and also of society as a whole. 

If there are differing and at some points logically conflicting 
normative patterns applicable to behavior in different areas of the 
society, in principle careful scheduling and contextual segregation 
might serve to handle potentially disruptive consequences: a time 
and place for everything, and everything in its time and place. 
Such mechanisms do indeed help to accommodate potentially con- 
flicting substructures to one another. Parsons reasons, but by them- 
selves they are never quite sujBBcient. For one thing, concrete 
individuals always participate as role players in many different 

36 • Edward C. Devereux, Jr. 

relational systems, and the equilibrium and integrity of the per- 
sonality is threatened by too gross and rapid shifts in normative 
contexts. It is not just a matter of changing hats to move from 
one role to another, particularly if the normative patterns have 
become internalized as integral components of personality. For 
another thing, contexts are never completely segregated; there are 
always ambiguous situations where one is unsure which pattern 
of norms should apply. And besides, there is what Parsons calls 
the strain toward consistency at the level of cultural systems them- 
selves. No ethic worth its salt will long remain a hodgepodge of 
particularistic rules, each tailored to its own special context: it 
will respond to pressures toward codification and generalization. 
But in doing so, while it may appeal more forcefully to the loyalty 
and support of its adherents, it may also overstep its own bounds 
and augment potentialities of conflict with alternative systems in 
the same society. In effect, Parsons argues, man's need for at least 
the appearance of normative consistency creates problems which, 
at a strictly operative, functional level, might not need to be faced 
at all. But man puts them there, and so face them he must. 

What happens, as Parsons sees it, is something like this. Every 
society tends to establish at least one core of common ultimate 
values which serves an important unifying function. But this core 
value system, however splendid it may seem in principle, or on 
Sunday mornings and the Fourth of July, can never be fully opera- 
tive at all concrete levels of social structure. Its norms are always 
a bit too vague and general to apply to all the concrete situations 
which need to be defined on Tuesday afternoons. And so each 
sub-system and situational context tends to develop its own special 
normative patterns. Parsons argues that because of the heterogeneity 
of functions which need to be served in a differentiated society, 
it is manifestly impossible for the norms of all of these sub-systems 
to form a single, logically coherent and consistent system. Yet 
neither can they afford to be flagrantly incompatible. 

At the level of any particular sub-system, there is thus the addi- 
tional imperative of normative compatibility. The norms which 
regulate relations in the U.S. Navy cannot be merely those appro- 
priate for the instrumental functions of a military organization. 
They must yield at points to take into account the fact that this 

parsons' sociological theory • 37 

particular navy is called into being to serve a free, individualistic, 
nonauthoritarian, and democratic society. A somewhat different 
set of norms might develop in a military organization serving a 
totalitarian society. 

But the potentiality of conflict is still present, Parsons argues, 
and hence there is a need for a variety of mechanisms which will 
serve the function of accommodation: scheduling, symbolic and 
contextual segregation, rationalization, and other such devices 
serve to mask the points of conflict, insulate conflicting structures 
from head-on contact and provide at least an appearance of com- 
patibility. The integration of a society which depends too much 
on such devices, however, is evidently somewhat precarious. 

From the point of view of society as a whole, there is yet another 
problem. As Parsons sees it, certain core institutional structures 
develop to serve the major functional needs of the society, while 
maintaining at least a show of normative compatibility. But there 
are always certain residual needs which cannot be met legitimately 
without a direct affront to the dominant value systems. Such needs 
may reflect instinctive biological or psychological needs which 
have somehow got crowded out of recognition. More likely, they 
represent derivitives of strains in the dominant social structure itself, 
and their precipitates in personality. Yet if these needs remain 
unserved, strains will build up and perhaps discharge in socially 
disruptive channels. As if to forestall this dysfunctional outcome, 
society is thus forced to develop a set of adaptive structures which 
explicitly institutionalize sub rosa patterns of behavior which 
are deviant in terms of the dominant value system. Parsons would 
point to prostitution and gambling as examples of institutionalized 
deviant patterns. While such adaptive institutions may indeed solve 
some problems by providing channels of relatively safe release for 
potentially disruptive energies, it is clear that they also raise still 
others with respect to the over-all integration and stability of society. 

Finally, there are many tricky problems of integration, and hence 
potential sources of conflict and strain, in the relationships of 
various sub-systems within a given society to one another. Family 
institutions and economic institutions in our society serve rather 
different functions. But not just any type of family and any type of 
economic system can coexist in the same empirical society. Parsons 

38 • Edward C. Devereux, Jr. 

has argued that a famihstic system such as that of classical China 
would be drastically dysfunctional in an industrial, capitalistic 
society such as our own. In effect, commitments made in one area 
of the social structure restrict alternatives in others. In addition to 
the universal imperatives which must somehow be met by all 
societies, there are structural imperatives peculiar to each specific 
type of society, imperatives relevant to the structural compatibility 
and mutual articulation of various sub-systems in the same society 
with one another. 

But perhaps that is sufficient to show why Parsons regards societal 
integration and equilibrium, not as something to be taken for 
granted, but rather as something posing a major challenge for the 
social analyst. I shall pursue this problem further a bit later on. 
But first, I must consider how Parsonian theory deals with the 
problem of structural differentiation. 


We have been dealing so far with what are essentially uni- 
versals in Parsonian theory. We have talked about certain generic 
properties of action and interaction, and of personality, social, and 
cultural systems as Parsons conceives them; and we have said a bit 
about the kinds of relationships he sees among them. In effect, 
we have established a frame of reference rather than a substantive 
theory. We have the beginnings of a conceptual scheme, but no 
variables. The frame of reference and the conceptual scheme, how- 
ever, should provide a useful guide for pointing out the kinds of 
phenomena for which suitable variables will need to be developed. 
The U.S. Navy and the Jones family are both examples of social 
systems, and what has been said so far of social systems will apply 
equally to both. But we may suspect that the differences between 
these systems are quite as important as the similarities, and if we 
are to make any progress in developing a useful substantive theory, 
we had better learn to talk about these too. But how? How does 
one go about comparing the relationships between captain and 

parsons' sociological theory • 39 

crew with those between husband and wife? How does the doctor- 
patient relationship compare with that of salesman-customer? 

In seeking appropriate variables for his theoretical system, Par- 
sons was guided by three principal criteria. First, the variables 
should be completely general and permit comparisons between 
groups of any sort whatever and across cultures. The special vocab- 
ularies which have been developed for describing particular kinds 
of social systems, for example, family systems or economic systems, 
will not do. Indeed, such vocabularies have often functioned as 
blinders to impede the development of general theory: although 
the family is undoubtedly a small group, very few family sociolo- 
gists have made any systematic eflFort to draw upon the rich funds 
of theory and research developed by small-group sociologists. 

Second, the variables should be relevant for the action frame of 
reference. For Parsons, this means that when applied to particular 
actors they should yield a classification of types of orientations, 
when applied to social systems they should serve to classify role 
expectations, and when applied to cultural systems they should deal 
with types of normative patterns. Moreover, because of the inter- 
penetrating character of these orders of systems, the same set of 
variables should serve to deal with all three. 

Finally, the variables should be relevant for the analysis of the 
functional problems about which system differentiation takes place. 
If a business firm and a mental hospital have somewhat different 
forms of organization, it is probably because these organizations 
must serve different functions. Not all categories referring to ob- 
served differences between these organization types will do. The 
variables selected should hit upon points of similarity and differ- 
ence crucial to the functioning of the system: you should be able 
to demonstrate that a change in state of any one of the variables 
would have some important consequence in terms of system func- 

The outcome of Parsons' thinking about these matters was the 
now-famous set of pattern variables, long the hallmark of Parsonian 
theory and regarded by some as his most important single theo- 
retical formulation. These were a set of five dichotomous variables 
conceived as constituting universal and basic dilemmas confronting 
any actor in any social situation. Parsons argued that each variable 

40 • Edward C. Devereux, Jr. 

represented a fundamental problem of orientation which the actor 
would somehow have to resolve either one way or the other; more- 
over, he would have to come to terms with all five before arriving 
at any determinate orientation. 

Let us consider how these dilemmas were conceived. 

1. Affectivittj -Affective neittrality. Originally Parsons used this 
variable simply to characterize one attitudinal component in social 
relationships— whether aflFectivity were present or absent. Relation- 
ships between husbands and wives in our society typically incorpo- 
rate a high level of affectivity, whereas those between social worker 
and client do not. Nothing was implied about whether the affect 
was positive or negative, and obviously there are wide variations 
from case to case: some marriages go dead, affectively speaking. 

But Parsons was never interested in particular cases as much as 
in general norms. In terms of the institutionalized norms relating 
to role expectations, American marriages are expected to involve 
affectivity, while relationships between social workers and clients 
are not. In each case, appropriate sanctions are present to deal 
with deviations. And this in turn implies for Parsons that these 
institutionalized definitions are probably appropriate for the func- 
tioning of these two sorts of relationships: if the social worker, or 
doctor, allows himself to become affectively involved, he may get 
distracted from the task at hand. 

In his more recent work with this variable. Parsons has been 
particularly concerned with these functional consequences. He came 
to see affectivity vs. neutrality essentially as the dilemma of accept- 
ing immediate gratification from the situation at hand or of defer- 
ring gratification and accepting discipline. The latter alternative, 
he argued, is usually involved in instrumental or task-oriented situa- 
tions: the situation or relationship is not to be enjoyed in its own 
right but to be evaluated and used. 

2. Specificitif-Diffuseness. This dichotomy points to yet another 
attitudinal dimension in orientation toward social objects— what 
Parsons refers to as the scope or inclusiveness of the relationship. 
The marriage relationship may serve as a prototype of the diffuse 
relationship: ego orients to alter as a total personality. At the 
opposite pole one might point to the highly specific character of 
the relationship between clerk and customer in a drugstore: the 

parsons' sociological theory • 41 

parties are brought together for a specific and hmited purpose, and 
at least at the level of institutionalized expectations, all other aspects 
of the personalities of the role-players may be ignored. That these 
expectations are not always observed in practice is evidenced by 
the standard deviation of the boss-secretary relationship, as end- 
lessly depicted in the pages of Esquire during the past twenty 
years. But this is deviation. Parsons would insist; there are sanc- 
tions and probably also dysfunctional consequences. It would 
presumably be equally deviant and dysfunctional for the husband 
to treat his wife in the functionally specific role of housekeeper. 

Although there are evidently wide variations in scope appro- 
priate for different sorts of relationships, Parsons sees the basic 
dilemma as essentially dichotomous. In a functionally specific rela- 
tionship, if alter demands more of ego than ego is prepared to give, 
the burden of proof is upon alter to show why this demand is 
justified. The prototype is the contractual relationship: if something 
is not there in writing, you have no right to claim it. But where 
diffuseness is the norm, as in the family, the burden of proof would 
fall upon ego to demonstrate that some still higher claim prevents 
his compliance. 

3. Universalism-Particularism. This is the first of two dilemmas 
which, in Parsons' system, pertain to modes of categorizing social 
objects. Shall the object be judged in terms of some universal or 
general frame of reference, or in terms of some particular reference 
scheme in which ego is himself personally involved? Whether some- 
one is a good doctor, a competent secretary, or a beautiful woman 
are presumably matters to be determined on universalistic grounds. 
But while certain modes of behavior might be evoked toward 
beautiful women or deserving children in general, where one's own 
wife or child is involved, one is committed in many special ways, 
regardless of beauty or desert. Parsons would point to nepotism as 
prototypical of situations in which particularistic criteria are given 
precedence over universalistic ones; and he would argue that in 
many instrumental social systems an intrusion of nepotism would 
be dysfunctional. One should choose a doctor on the basis of his 
competence and not on the ground that he is friend, neighbor, or 
cousin. Essentially the dilemma is whether cognitive or cathectic 
criteria should take precedence in defining the relationship. 

42 • Edward C. Devereux, Jr. 

4. Quality-Performance. The second mode of characterizing social 
objects represents an attempt by Parsons to restate in more general 
terms the central issue in Linton's ascription-achievement dichot- 
omy. This central dilemma, as Parsons conceives it, turns upon 
whether the primary consideration, in defining a relationship, is 
given to some ascriptive quality of the object— age, sex, beauty, 
possessions, status, and so on— or to some particular complex of 
performances. What matters most: who or what the person is, or 
what he has done or can be expected to do? In preparing to scold 
a lady driver, do you address your rernarks primarily to the lady or 
to the driver? How you behave will depend upon which side of 
this polarity gains primacy. 

5. Self-orientation-collectivity-orientation. In his study of the 
medical profession. Parsons was struck with the observation that 
the relationship between doctor and patient is expected to be dis- 
interested, while that of salesman and client is expected to be 
self-interested. This is evidently not a question of diflFerences in 
personality or personal motivation. It is rather a matter of institu- 
tional regulation: whatever his personal motivations, the business- 
man is expected to make decisions with his eye primarily on the 
balance sheet of his own firm, whereas the doctor is expected to 
think primarily of the welfare of the patient, placing his own 
economic welfare in second priority. One can easily demonstrate, 
Parsons argued, that a norm of caveat emptor would be drastically 
dysfunctional for the medical profession. 

Attempting to cast this problem in more general terms, it seemed 
to Parsons that, in some relationships, what are essentially moral 
considerations are expected to be given primacy, whereas in others 
they are not. And what is morality, he reasoned, but the claim of 
some superordinate collectivity upon the individual or sub-collec- 
tivity? The problem becomes a crucial one for the relationships 
among different orders of systems. Does the husband-father act 
primarily for himself or for his family as a whole? Does the depart- 
mental executive in a business firm act primarily in terms of his 
own personal interests, in terms of the interests and welfare of the 
department he serves, in terms of the interests of the firm as a 
whole, or in terms of some still broader collectivity, perhaps of 
society in general? It should be clear that precisely the same 

parsons' sociological theory • 43 

behavior which is collectivity-oriented in terms of the individual or 
some sub-collectivity may yet be self-oriented in terms of the 
larger system referent. The dedicated businessman may selflessly 
serve a corporation which is essentially self-oriented in its dealings 
with surrounding systems. 

These five dichotomies, then, represent for Parsons the universal 
dilemmas of orientation which must somehow be resolved before 
any determinant orientation is achieved. He has argued, moreover, 
that at this level of generality, the list is apparently exhaustive. 
There are these five and no others.* Since these variables are con- 
ceived as being analytically independent of one another, the set 
thus generates, by cross tabulation, a typology of thirty-two logically 
possible patterns of orientation. But in this theoretically-derived 
typology, Parsons has observed, there are many empty cells: em- 
pirically, certain combinations of variables tend to cluster whereas 
others never occur, and probably represent empirical impossibilities. 

The pattern variables were described as representing dilemmas 
faced by individual actors in attempting to define social situations. 
But what determines which pattern will be selected? In principle, 
it might be wholly a matter of free individual preference, rooted 
in the personality structure. Are some types of individuals, one 
might ask, predisposed to define situations particularistically, others 
to define situations universalistically? Although Parsons is willing 
to entertain this possibility, he is not much interested in it. For him 
the more important fact is that these choices tend to be defined by 
the culture and institutionalized in the form of normative patterns 
held to be appropriate for different types of relational systems. If 
the doctor's wife has planned a special dinner party and if an 
emergency call comes just as the guests are arriving, the culture 
is clear about how the doctor ought to resolve the dilemma. The 
test of an institutionalized expectation, Parsons reasons, is precisely 
this notion of oughtness and the presence of sanctions. In this case 

* In one recent discussion of this position, Parsons entertained the 
notion that perhaps a sixth dichotomy— long-run versus short-run 
focus of vakiation— might be needed to make the fist exhaustive. But 
in his subsequent work he has made no attempt to employ this addi- 
tional variable. See "Some Comments on the State of the General 
Theory of Action," American Sociological Review, XVIII, No. 18 
(Dec. 1953), pp. 618-631. 

44 • Edward C. Devereux, Jr. 

the claims of the larger collectivity are expected to take precedence 
over those of the family. 

By applying this test— on which side does culture tend to throw- 
its weight in a tight decision?— Parsons is able to describe in pattern- 
variable terms the profiles which characterize many different sorts 
of relational systems. Relationships in families and friendship groups 
typically display a pattern characterized by affectivity, diffuseness, 
particularism, quality orientation, and collectivity orientation. In 
contrast, the relationships between business firms and customers 
in our society tend to stress precisely the opposite pattern: affec- 
tive neutrality, specificity, universalism, performance, and self- 
orientation. These two polar cases serve to represent, in Parsonian 
terminology, the ideal-typical patterns of Gemeinschaft and 
Gesellschaft respectively. But with the additional leverage provided 
by his system of pattern variables. Parsons is also able to describe 
other, more complicated intermediate types. Relationships between 
doctors and patients, he demonstrates, are like those of business 
firms and customers except for one crucial difference: they are 
expected to be collectivity-oriented. Relationships between social 
workers and clients are like those of doctors and patients, again 
with one crucial difference: they tend to be more diffuse in scope. 
By some such procedure as this, Parsons has attempted to describe, 
classify, and compare the structures of a wide variety of relational 


The pattern variables have provided for Parsons a conven- 
ient tool for the description, classification, and systematic comparison 
of social structures. But structural description has never been his 
primary goal. If there are these observed differeneces in relational 
patterns, the important questions for Parsons have always been: 
Why? What are the bases on which such structural differentiation 
occurs? What differences do these differences make? It should be 
clear that by themselves the pattern variables do not provide 
answers to these questions. They were selected in such a way, how- 

parsons' sociological theory • 45 

ever, that they should be relevant for the sort of answers Parsons 
has been seeking. 

The basic form of Parsons' answer is fairly clear: the normative 
pattern which becomes institutionalized for any particular type of 
relational system will tend to be one which is somehow relevant 
for the eflFective functioning of that type of system. Families must 
perform a set of functions diflFerent from those of business firms, 
and the normative patterns which govern these difiFerent types of 
institutions will reflect these differences in function. 

In attempting to test the goodness of fit of any particular norma- 
tive pattern for the relational system it serves, Parsons employs a 
variety of devices. Most frequently he seems to depend upon a 
kind of "Gedankenexperiment" : he simply asks, what would be the 
consequences for the system under study of some imagined devia- 
tion from the established normative pattern? What would happen 
to the doctor-patient relationship, for example, if this relationship 
were allowed to become diffuse, or particularistic, or self-oriented? 
Drawing upon funded knowledge, Verstehen, and careful reasoning, 
he attempts to demonstrate that any such departures from the 
established pattern would have seriously disruptive, that is dys- 
functional, consequences for the system in question. 

But Parsons does not have to depend wholly upon his imagina- 
tion in carrying out this analysis, for nature provides a variety of 
experiments he is able to exploit. Much may be learned from 
observed instances of actual deviation from the established pattern. 
Systematic comparison of relational systems which are similar in 
some respects but different in others also provides a point of 
leverage. By a careful and imaginative use of such comparisons, 
Parsons attempts to come to grips with the unique combinations 
of circumstances which give rise to the particular normative pat- 
tern under study. Thus in the course of his analysis of modern 
medical practice, we find him exploring the points of similarity 
and difference between various aspects of medicine and an aston- 
ishing number of other somewhat parallel phenomena. To mention 
only a few of them: he compares medicine with magic; medical 
science with military science (both involve high stakes and large 
elements of risk and uncertainty); the doctor with the engineer 
(both play technical roles, but the latter deals with nonhuman 

46 • Edward C. Devereux, Jr. 

materials which do not have emotional reactions); the doctor with 
the wise man ( the latter has a more generalized wisdom and 
authority); the doctor with the priest (both deal with death, but 
the latter is more concerned with its sacred aspects); the doctor 
with the army officer (the authority of the latter is backed by 
coercive sanctions). Each such comparison provides Parsons with 
a point of leverage for analyzing some special feature of the 
doctor's role. His principal paper on this subject (SS ch. X) con- 
tains, by actual count, no less than thirty such comparative refer- 
ences. Some of them are truly astonishing: who but Parsons would 
ever stop to wonder why it is that a child typically has two parents, 
while a patient has only one therapist? Is it something about Par- 
sons' theory, or simply his lively imagination, that leads him to 
raise such questions? 

The form which Parsons follows in his empirical analysis is 
extremely complex. Before attempting to state in general terms 
just what it is he seems to do, we had better take a look at a 
typical sample of Parsonian argument. Let us jump into the middle 
of his analysis of the doctor-patient relationship (SS ch. X) and 
try to follow through just one thread in the complex web. It seems 
to run something like this: 

1. In order to perform his technical functions adequately, the 
doctor must have access to the body of the patient, and also to 
certain areas of private information about the patient. (A tech- 
nical imperative is established.) 

2. In other stiuctures ia which both the doctor and tlie pa- 
tient are involved, body access and private information are 
severely taboo'd and occur primarily in a nexus of intimate 
friendship or marriage. (The problem of multiple roles and con- 
flicting normative standards is cited.) 

3. If the doctor is defined as a non-intimate or stranger, the 
patient may feel resistance to revealing secrets or allowing body 
access; but this withholding would be dysfunctional for the 
technical performances of the doctor. (A source of strain creates 
a functional problem which must somehow be resolved.) 

4. Caught up in this confusion of symbolic meanings, the pa- 
tient may attempt to resolve the problem by trying to assimilate 
the doctor to a nexus of intimate personal relationship, perhaps 
seeking "secondary gains" from this area of permissive intimacy. 
(A tendency toward deviance is established.) 

parsons' sociological theory • 47 

5. If the doctor were to allow himself to be drawn into an 
emotionally-charged personal relationship with the patient, the 
attitude of scientific objectivity essential for the rational treat- 
ment of the patient as a "case" might be seriously hampered. 
(The threatened deviance would be dysfunctional for the tech- 
nical role performance of the doctor.) 

6. Proper therapy, in psychiatric cases, also requires that the 
doctor should not reciprocate the attachments and attacks of the 
patient. {An additional technical imperative is cited.) 

7. If the doctor were to become involved in personal relation- 
ships with his patients, his relationships with his own wife and 
friends might also be jeopardized. {The threatened deviance 
would he dysfunctional for other system-referents as well.) 

8. In order to handle these sources of strain and deviance, 
block these potential dysfunctions, and support the technically 
required functioning of the doctor-patient relationship, an appro- 
priate set of mechanisms is necessary. These must be of such a 
nature as to permit functionally-relevant body access and com- 
munication, without undue discomfort or embarrassment and 
without dysfunctional side-effects. (A structural imperative is 
defined. ) 

9. The institutionalized patterning of the doctor-patient rela- 
tionships in terms of the norms of functional specificity and 
aflFective neutrality helps to keep this relationship on its func- 
tionally-required track and to mitigate the strains and dysfunc- 
tional side-effects to which it is subject. 

(a) The norm of functional specificity serves to define and 
restrict the doctor's access to privileged information and con- 
tact, in terms of a criterion of technical relevance, and this 
restriction functions to allay the anxieties of the patient about 
the possible consequences of such privileges. 

(b) The norm of afiFective neutrality defines the expected 
attitudes within these limits: keeping the relationship "pro- 
fessional" and affectively neutral serves to protect both parties 
from inappropriate and potentially dangerous involvements, 
and peiTnits the doctor to give technical considerations his 
full attention. 

{The established normative pattern is shown to he functionally 

10. But institutionalized normative patterns cannot be ejBFec- 
tive unless they are communicated, internalized, backed by ap- 
propriate sanctions, and bolstered by appropriate symbols. {Still 
another sort of derived structural imperative is invoked. ) 

11. In the doctor-patient relationship the norms of functional 
specificity and affective neutrality are communicated and bol- 

48 • Edward C. Devereux, Jr. 

stered by a variety of control mechanisms: the symbohsm of the 
"doctor's office," the white coat, the presence of the nurse 
"chaperone," the framed medical degree, the scientific-looking 
apparatus, and so on. These contextual arrangements function to 
remind both doctor and patient of the roles they are expected to 
play. (Appropriate mechanisms of social control are shown to 
serve structural imperatives.) 

Perhaps that is sufficient to give some impression of the line of 
argument Parsons pursues in carrying out what he describes as 
dynamic structural-functional analysis. It should be clear that 
what we have seen is just one small fragment of the total argu- 
ment, and a fragment necessarily leaves far too many things dan- 
gling. Yet the whole point of a general theory and of a systematic 
analysis made in terms of it is that nothing at all should be left 
dangling. In the original web from which we have abstracted this 
single thread, Parsons begins by examining such issues as these: 
In what specific ways is illness dysfunctional for society? Under 
what combinations of circumstances does the treatment of disease 
come into the province of science and get removed from the tradi- 
tional contexts of religion and magic? What are the special socio- 
logical characteristics of the "sick role" in our own society? Before 
he is done, he has boxed the compass three times over, wdth the 
result that the argument does indeed seem to achieve some sort 
of closure. Piece by piece the loose ends are somehow picked up 
and tucked back in; the degrees of freedom are gradually whittled 
away until in the end it appears that the entire system is over- 
determined and locked shut. Knowing what we do of Parsons' views 
regarding the precarious nature of social equilibrium, however, 
we cannot believe that he really expects things to stay at rest 
very long. 

Considering this example, together with many others in which 
Parsons has worked his way through the analysis of various empiri- 
cal phenomena, let us attempt to state in more general terms the 
principal elements in his formula for dynamic structural-functional 

First, there is an analysis of a set of needs or imperatives. These 
may be needs of individuals, as biological or psychological systems, 
or they may be needs of particular social or cultural systems, or of 


society as a whole. A general analysis will necessarily touch upon 
the needs of all of these different sorts of system-referents. The 
formula for establishing a need seems to be fairly clear: you 
attempt to demonstrate that if the need is not somehow dealt with, 
there will be dysfunctional consequences for the system in ques- 
tion. A consequence is considered dysfunctional if it disturbs the 
equilibrium of the system beyond some normal range of tolerance. 

The range of phenomena treated as needs in Parsonian analyses 
is vast, but there does appear to be a certain order among them. 
The most important are those which he regards as universal, in 
the sense that they stem from the more or less fixed parameters of 
heredity and environment, and the limits these impose upon any 
social system. Given the facts of biology, the organism must eat 
to survive. Given the helplessness of the human infant, some pro- 
vision must be made for its care and training. Given the cycle 
of life and death, societies must contrive to replace their members. 
Parsons refers to needs of this sort as universal imperatives or 
functional prerequisites, and his list of them is relatively small. 

All the rest of the needs which figure in his analyses are regarded 
as secondary, derived or contingent. They stem from circumstances 
peculiar to some particular type or state of a given system. It is 
a universal imperative for any society to make some provision for 
its food supply. But if some particular society, given its own 
peculiar environment and state of technology, must do so by con- 
triving to catch large fish at sea, at once a variety of other impera- 
tives come into being. There must be boats and nets, and men to 
man them; there must be coordination of activities and hence 
leadership; there must be some definition of who owns what, and 
of how the catch will be distributed; there must be some provision 
that those who cannot fish may still be fed; and there must be 
some provision that other vital interests of the society will not 
all be neglected while the men are off in the boats. 

It is of course this proliferation of derived needs and their com- 
plex interconnections which provide for Parsons the challenge he 
most enjoys. There appear to be three principal varieties of such 
nonuniversal imperatives which figure in Parsonian analyses: there 
are the technical or instrumental imperatives, that is those which 
arise in connection with bringing instrumental processes to bear 

50 • Edward C. Devereux, Jr. 

on some intermediate goal or objective; there are the organizational 
or structural imperatives, that is those which are concerned with 
the establishment and maintenance of any particular form of social 
organization; and there are the imperatives of compatibility, that 
is those which are concerned with the articulated adjustments of 
the diflFerent sorts of systems in any particular society to one another. 

It should be clear that all of these derived needs are relative 
to the equilibrium or survival of some, particular form of a system, 
and that they are not necessarily involved in the survival of the 
society as a whole. One could easily make a listing of imperatives 
appropriate for the survival of the pattern of racial discrimination 
in the South, without implying that a continuation of this pattern 
is essential for societal survival. 

Second in the Parsonian plan of structural-functional analysis 
is a detailed account of the structures, processes, and mechanisms 
through which these various imperatives are served. The advantage 
of the functional point of view, Parsons maintains, is that it sup- 
plies a continuous criterion of relevance for structural analysis; the 
insistent question is always this: what are the consequences of 
this or that particular process or structural element for the various 
needs of the systems it serves? What would happen if this particu- 
lar component were absent or altered in some way? How Parsons 
proceeds with this phase of the analysis has been illustrated before. 
It should be clear that this step of the analysis is concerned with 
consequences, and not with origins. It is not essential to the argu- 
ment to say that these structures have come into being to serve 
these needs, although some such teleological postulate is often 
implicit in structural-functional analysis, including some of those 
by Parsons himself. 

The matching of structures, functions, and needs involves some 
rather slippery problems for the analyst. For one thing, as soon as 
any particular structural form comes into being to serve a given 
need, a host of additional, derived imperatives spring up in con- 
nection with the maintenance of this particular structural form. 
For another, it is clear that any operative structure has conse- 
quences or outputs with respect to a variety of different system- 
referents; while it may serve the needs of some of these most 
admirably, it may produce dysfunctional consequences for others. 


parsons' sociological theory • 51 

Again, it is evident that almost any particular need is usually 
served by a variety of different structural components. Consider a 
simple-minded example: clothing may serve the needs of warmth, 
modesty, status differentiation, and sexual attraction. Yet each of 
these needs is also served by other devices as well— status differ- 
entiation, for example, by language, possessions, style of life, and 
so on. And of course the nature of the "need" for status differentia- 
tion varies for different system-referents in the same societv, and 
varies widely from one type of society to another. 

It is this complex intermeshing of needs, structures, and func- 
tions which makes it essential that the analysis be systematic and 
complete. It is not enough to cite a need and point to a structural 
form which serves it, or to propose an alternative which might serve 
it better. One must consider as well all of the collateral consequences 
both of the observed pattern and the proposed alternative for 
other needs of the same and other system-referents. And one must 
continually appraise the needs themselves, keeping in mind the 
contingent character of most of them and the chains of circum- 
stance from which they derive. The final outcome of a systematic 
structural-functional analysis would be, ideally, some kind of inven- 
tory in which all of the consequences of existing arrangements and 
contemplated alternatives were projected against a carefully speci- 
fied hierarchy of needs and values, to the end of arriving at some 
over-all balance sheet of net gains or losses for society as a whole 
or for the sub-system under examination. 

There is another outcome, however, in which Parsons seems to 
be somewhat more interested— the assessment of the systems under 
study in terms of the problems of internal dynamics and equilib- 
rium. His procedure here is of some interest. Having described the 
structural circuits of the phenomena under study, he turns on the 
motivational currents and seeks to observe what happens when 
the juices of affect are coursing through them. He tries to account 
for the forces which tend to generate the system and hold it in a 
steady state. But he also looks for the points of strain or tension, 
and for the forces which tend to pull the system apart. Having 
located these, he looks for mechanisms of control which function 
to buttress the system at strain points or otherwise to restore the 
balance. He goes on to ask whether any strains remain unresolved 


52 • Edward C. Devereux, Jr. 

and, when he finds them, probes for adaptive structures which may 
have sprung up to serve them. 

Parsons' approach to the problems of structural-functional anal- 
ysis obviously leans rather heavily on the prior work of Durkheim 
and Malinowski. Indeed, it appears to this reviewer that Parsons 
has never produced as direct a statement of what is involved in 
this type of analysis as that proposed by Malinowski some thirty 
years ago.* Yet in Parsons' hands the method has grown in richness 
and complexity, perhaps simply because of his skill at keeping an 
extraordinary number of analytical balls in the air at the same 
time, perhaps because of his almost fanatical drive toward sys- 
tematic completeness. Parsons' special contribution to structural- 
functional analysis, it seems to me, lies in the area of what he 
calls dynamic analysis. More than most others who have worked 
this field. Parsons has attempted to draw in, at every step of the 
path, the relevant psychological and motivational factors. And it is 
these, of course, which provide the forces, strains, and tensions 
with which he likes to deal. 

I have commented earlier on Parsons' concern with the problem 
of equilibrium, and argued that this is not something he takes 
lightly for granted. Actually, I suspect that Parsons is not so much 
interested in the final product, equilibrium, as he is in the processes 
which bring this about. For him, the phenomenon of equilibrium 
serves as an heuristically useful dependent variable or criterion of 
effect, in terms of which the manifold processes of system function- 
ing may be analyzed. It supplies an insistent standard of relevance 
for every step in the analysis. The fact that a social system survives 
in its environment, despite many instigations to change or deviance, 
indicates that it has somehow managed to cope with its complex 
problems and needs. The heart of the analysis lies in specifying 
the needs and the mechanisms through which they are served, and 
then of attempting to arrive at some notion of the over-all balance 
of forces coursing through the system. The criterion of dynamic 
equilibrium serves as a sort of summation function for this analysis. 

* For Malinowski's best brief statement, see "Culture," Encyclopaedia 
of the Social Sciences, IV (New York: The Macmillan Company, 
1930), pp. 621-645. 

parsons' sociological theory • 53 


The pattern variables and the problems of structural-func- 
tional analysis occupied a central position in Parsons' thinking over 
a long period of time. In his most recent works, however, he appears 
to be moving away from them. He has come more and more to 
regard the concept of structure as a sort of analytical half-way 
house. Structure, as Parsons sees it, represents at best a convenient 
way of codifying and talking about certain apparent constancies in 
social phenomena, before their internal processes and dynamic 
laws are fully understood. But when we make structure our primary 
focus of attention, he has reasoned, there is a danger that we will 
somehow reify it and bypass more basic questions about the 
processes that generate and maintain these apparent constancies. 
Structure presupposes frozen process; but in reality, process never 

Consider in this connection what is implied by Parsons' concep- 
tion of dynamic equilibrium. The equilibrium of a social system, 
he observes, is of a rather different sort than that of a table or a 
pile of sand, which stay as they are simply because nothing is going 
on within or outside of them to provide an impulse for change: 
the principle of inertia alone is sufficient to account for this sort 
of static equilibrium. With social systems, as with organic systems, 
Parsons argues, such static equilibrium is never possible, for two 
reasons: there is always a certain amount of continuing process 
within the system which provides an impulse for change of state; 
and there is always an element of flux in the external situation which 
tends to throw the system off balance. Dynai ic equilibrium is 
not so much a matter of a system's remaining always in a steady 
state as it is of the system's having the capacity to return to some 
status quo ante after each minor disturbance, by means of appro- 
priate adjustments. To refer to processes which serve this equili- 
brating function, Parsons uses the term mechanisms: organic 
systems have regulative mechanisms, personalities have mechanisms 
of defense, and social systems have mechanisms of social control. 

54 • Edward C. Devereux, Jr. 

Although some such model as this has been in Parsons' thinking 
all along, in the last decade he has been pursuing some of its 
implications much more intensively, primarily, it would appear, 
as a result of his close association and collaboration with Robert 
F. Bales. 

Bales' concern with the observational study of interaction process 
in small face-to-face groups led him to work out a system of 
categories for classifying everything Which was said or done, at 
the moment it occurred. These categories, twelve in all, were 
designed to be relevant for what Bales conceived to be the princi- 
pal functional problems which would need to be faced by any 
small problem-solving group. Essentially, Bales reasoned, these 
would be of two principal types : those concerned with the solution 
of the problems imposed by the task itself, and those concerned 
with the motivations of group members and the establishment of 
a sufficient level of cohesion to permit the group to function as a 
unit in dealing with its task. Six of the categories fell into the task 
area and six into what Bales called the social-emotional area. 
Within each set, three represented forms of positive interaction 
and three of negative interaction.* 

Although it may be used for many other purposes as well, the 
Bales technique was particularly well suited for studying sequences 
of group processes, along a time line. Set a group to interacting 
about some problem, record what is said and done in terms of the 
twelve categories, and then compare what happens in each suc- 
cessive phase of the interaction. Repeat this process with a large 
number of diflFerent groups and see whether there is any consistent 
sequence of activities through which groups typically move while 
solving a problem. With respect to the categories in the task area, 
Bales had expected, on the basis of his theory of problem-solving, 
that in the initial phase interaction would center largely about the 
problems of orientation, that in the next phase evaluation would 
become the main focus of interest, and that in the final phase, 
problems of control would receive the major attention of the group. 
With respect to the categories in the social-emotional area, he had 

* Robert F. Bales, Interaction Process Analysis (Cambridge, Mass.: 
Addison-Wesley Press, 1950). 


parsons' sociological theory • 55 

expected that these would become increasingly prominent, both in 
their positive and negative forms, as the groups moved from one 
phase to the next. His findings tended in general to confirm these 
expectations.* There was thus clear empirical evidence that inter- 
action processes in small groups tended to be differentiated along 
a time line, and that this differentiation tended to follow a sequence 
of phases relevant for the functional problems of the group. 

The Bales technique also provided some fresh evidence about 
the bases of role differentiation in small problem-solving groups. 
By making a separate record for each member of the group, a 
series of profiles may be obtained showing the categories in which 
the contributions of each most frequently fall. By this device. Bales 
and his associates were able to demonstrate that there are indeed 
consistent differences in the roles played by different members of 
the group. These differences emerge fairly early, tend to stabilize, 
and to carry forward from one meeting to the next. And on what 
basis does this differentiation occur? Among others, on the basis 
of leadership type. In many of the groups studied there were clear 
evidences of the emergence of a dual-leadership pattern, with one 
person assuming the role of task leader, another the role of social- 
emotional leader of the group. f 

It should be clear that the groups studied by Bales and his 
associates were not natural groups which existed as regular com- 
ponents of some larger sociocultural system. They were contrived 
groups, set up ad hoc for the purposes of the experiment, typically 
consisting of members who were strangers at the start and stood 
in no established relations to one another. The problems were also 
contrived and were not of the sort in which members of the groups 
might have had some prior involvement. So far as possible, Bales 
was interested in catching interaction process and the emergence 
of differentiated group structures in a sociocultural vacuum. The 

* R. F. Bales and F. L. Strodtbeck, "Phases in Group Problem- 
Solving," The Journal of Abnormal and Social Psychology, XLVI 
(1951), pp. 485-495. 

I See R. F. Bales and Philip E. Slater, "Role Differentiation in Small 
Decision-Making Groups," in FSI, ch. 5, and Philip E. Slater, "Role 
Differentiation in Small Groups," in Paul Hare, E. F. Borgatta, and 
R. F. Bales, eds., Small Groups: Studies in Social Interaction (New 
York: Alfred A. Knopf, 1955), pp. 498-515. 

56 • Edward C. Devereux, Jr. 

roles of which Bales talked were not "social roles" as Parsons con- 
ceived them, but merely behavioral roles. And whatever institutions 
or culture the groups possessed they appeared to have invented 
for themselves. 

For Parsons all this was heady medicine. Following in the grand 
tradition of Max Weber, he had devoted years to the study of 
comparative institutions, and had talked at length about the 
Protestant ethic, the American family, Japan, or Germany. To be 
sure, he had always tried to talk in terms of general analytical 
theory, but even so the fact remained that each object was histor- 
ically unique. Parsons' own insistence that the relevant sociocultural 
context be always taken into account threatened continually to 
ensnare him in the sort of cultural relativity and historicism of 
which he had accused the idealists. But now, in the challenging 
new formulations of Bales, he saw an opportunity to break through 
to a far more general plane of analysis without sacrificing anything 
essential to his own theoretical system. In any event, Parsons and 
Bales became active collaborators in two major monographs, the 
Working Papers (1953) and Family, Socialization, and Interaction 
Process (1955); and the resulting new directions have left their 
mark on all of Parsons' subsequent work. 

The core of this new synthesis consists of a reformulation of the 
functional problems faced by any social system whatever, large 
or small, institutionalized or contrived. These are now seen as four 
in number, two having to do with the relations of the system to 
the external environment, the other two with conditions internal 
to the system itself. 

First, there are the instrumental problems incident to goal attain- 
ment: these are seen as including the solution of relevant technical 
problems in terms of some means-ends schema and the coordina- 
tion of activities in such a way that the system moves toward 
whatever goals it has set for itself. 

Second, there are the problems of adaptation to the external situa- 
tion. As Parsons conceives it, adaptation is not merely the problem 
of coming to terms with the environment in whatever posture per- 
mits survival; it includes active manipulation of the environment, 
or of the system itself, to the end of acquiring facilities which have 
a generalized value as means for a variety of system goals. Capital 


accumulation, tool-building, and learning are all regarded as rele- 
vant for the adaptive problem. 

Third, there are the internal problems of integration: the focus 
here is upon the relations of units in the system to one another, 
and the problem that of establishing and maintaining a level of 
solidarity or cohesion among them sufficient to permit the system 
to function. 

Finally, there are the diflFerent but related problems of pattern 
maintenance and tension management. Both are concerned with 
conditions internal to the units themselves that nevertheless have 
consequences for system functioning. The problem of pattern 
maintenance is essentially that faced by an actor in reconciling the 
various norms and demands imposed by his participation in any 
particular social system with those of other systems in which he 
also participates, or with the more general norms of the broader 
culture. If there is serious role conflict or normative incompatabil- 
ity, the system will suffer the consequences. Tension management 
is defined as the problem of maintaining within the unit a level 
of motivational commitment sufficient for required role perform- 
ances. The notion here is that there are continuous changes of 
state within the units, with rise and fall of tension, and unless 
suitable measures are taken, these changes may potentially serve 
as instigation to deviance from the patterns established for the 

We may use a Navy destroyer as an example. Since the goal is 
finding and sinking enemy submarines, goal attainment consists 
of all of the activities and instrumental processes directly relevant 
for this task, as when the ship is at general quarters, with all hands 
at battle stations. Adaptation problems are those relevant for keep- 
ing the ship afloat and maneuverable in its sea-borne environment 
and in a state of readiness for whatever missions it may be called 
upon to serve. In addition to routine maintenance and drill at sea, 
periods in dry dock, recruitment policies, and training programs 
all serve the functions of adaptation. 

The internal components or units of the destroyer are its various 
departmental sub-systems— the departments of navigation, gunnery, 
engineering, communications, supply, and so on— and these in turn 
have individual officers and men as their units. The integrative 


58 • Edward C. Devereux, Jr. 

problem is essentially that of interdepartmental relations: keeping 
lines of authority and communications straight, coordinating the 
contributions of the various departments, and serving their needs 
in such a way as to mitigate interdepartmental jealousies and en- 
hance motivations to cooperate. Neat organization and high morale 
would appear to be the integrative goals. 

Finally, there are the two problems internal to the units them- 
selves which are relevant for system functioning. When depart- 
mental sub-systems are taken as units, the problem of pattern 
maintenance would appear to be this: what goes on or needs to 
go on within the departments in order to sustain a readiness to 
contribute to the performances required by the superordinate sys- 
tem? There is always some danger that component units will slip 
out of phase or out of field, as for example when any one depart- 
ment comes to regard its own goals as ends-in-themselves and not 
simply as instrumental contributions to the ultimate goals of the 
ship as a whole: an over-zealous maintenance department may 
produce some dysfunctional consequences for the ship as a fight- 
ing unit. Hence the need for adjustive mechanisms to keep these 
patterns in phase. The problem of tension management at the 
level of departmental sub-systems arises as a by-product of the 
fact that the flow of time and continuing process produces a con- 
tinual change of state within the sub-system: the men come on 
watch or go off on a liberty party, work, eat or sleep, strive for 
personal goals, compete, cooperate, or bicker. Even when the 
department is not directly contributing any output relevant for the 
broader system, these internal processes go on, tensions rise or 
fall, and hence unless they are somehow managed or controlled, 
there may be consequences dysfunctional for the broader system. 

When individual personalities are taken as the component units 
in the system, the problems of pattern maintenance and tension 
management are perhaps a little easier to visualize. Pattern main- 
tenance here involves the problem of the internalization of system 
goals and patterns, and the motivational commitment to them: for 
the individual officer or sailor, this will involve some efforts to 
reconcile these goals and standards with those of their other roles 
as husbands, church members, or citizens in a democratic society. 
Tension management here involves the problems which arise as 

parsons' sociological theory • 59 

a result of the continual changes of state which occur within the 
organism and personality system of the actor: if he is not eating, 
he is growing hungry; if he is working, he is growing tired; drives 
build up and need to be reduced, repressed urges demand to be 
dealt with. All these things occur not only while the actor is 
actively participating in his system role but also during periods of 
disengagement or latency. And evidently his success in dealing 
with them, in managing his tensions, and maintaining a posture of 
personality equilibrium or mental health, will have important con- 
sequences for his contributions to the system. Perhaps he can deal 
with them himself, by means of what Parsons calls his mechanisms 
of defense. But since it is also a problem for the social system, the 
system may need to provide some assistance. 

Let us now go on to mention briefly some of the uses Parsons 
and his various collaborators have been making of this analysis of 
the four functional problems of any social system. 

First, there has been a reformulation of the equilibrium problem 
in terms of a balance of phase movements (WP chs. 3, 4, 5). 
Parsons conceives of the four system problems as orthogonal 
dimensions in a sort of "action space" and argues that almost any 
concrete activity or process in which the system, or its components, 
engage will have some consequences for all of them. But since 
it is manifestly impossible to move in all directions at once, move- 
ment toward the "goal state" with respect to the solution of any 
one system problem may well involve movement away from the 
"goal state" with respect to the others. While the ship is in battle, 
maintenance problems get neglected. When the ship is in dry dock, 
attending to adaptive problems, it is for the while unable to fight. 
Leaves of absence for the sailors may relax tensions and restore 
mental health but raise new problems with respect to their rein- 
tegration in the ship's organization. There is no single state of a 
system. Parsons argues, constituting an optimum balance of gains 
and costs for all system problems simultaneously. The only solution 
to the equilibrium problem is a cycle of phase movements in which 
each type of system problem enjoys its moments of special attention. 

A typical sequence of phases might be one which starts with 
a focus on adaptive problems, the preparation of tools and facil- 
ities; goes on to goal attainment; when the work is done and the 

60 • Edward C. Devereux, Jr. 

goal attained, attends to the strains and tensions which its pursuit 
has entailed and utilizes the moments of gratification to re-establish 
feelings of solidarity' or cohesion; and then at last returns to a 
latent state in which component units find an opportunity to relax, 
blow oflF some steam, attend to private affairs, reconsider the 
premises of their involvement and prepare themselves for the next 
cycle of active participation. But Parsons does not insist on this 
particular order of phases. And in any event, he observes, which- 
ever phase is dominant at the moment, an eye must always be kept 
on the other problems as well, and if things get too badly out of 
line, something must be done at once. 

While Bales is busy plotting the cycle of phases during an hour 
of problem-solving by small experimental groups. Parsons' eye is 
already scanning the institutionalized social calendar. In a single 
day, the hours of eight to five are devoted to the instrumental prob- 
lems of adaptation and goal attainment, after work hours to the 
problems of integration, pattern maintenance, and latency. In the 
weekly cycle, weekends are reserved for special attention to these 
social-emotional problems. In the annual calendar, the same cyclical 
elements appear again, writ large: there are the work days, the 
holidays, and the holy days, each with their special functions. All 
these things, of course, are institutionalized and passed along as 
part of the content of the culture. But Parsons can now argue that 
their persistence is not merely a matter of blind learning. The 
evidence from Bales' experimental groups strongly suggests that 
if these important elements of temporal differentiation were some- 
how abolished men would quickly reinvent them. 

A second use Parsons has made of the system problems concerns 
role differentiation within social systems. We have noted before 
how Bales had observed the emergence of task leaders and social- 
emotional leaders in his experimental groups. Following his own 
characteristic train of thought, Parsons at once went on to inquire 
whether a similar basis of differentiation might not also be observed 
in established, institutionalized relational systems. Picking the family 
as a special case, he attempted to demonstrate that the husband- 
father is typically the specialist in the instrumental roles relative 
to the interactions of the family with the external environment, 



the wife-mother the speciaUst in the expressive or social-emotional 
areas concerned with relations internal to the family (FSI). 

The third area of application which Parsons found for the four- 
fold scheme of system problems lay in the reanalysis of the bases of 
structural difiFerentiation among social systems in the same society. 
So long as he had approached this problem from the concrete 
institutional level, he was in constant danger of becoming involved 
with historically unique situations and with fantastically elaborate 
schemes for trying to classify them. An approach from the direction 
of the four system problems oflFered a basis of classification which 
looked simpler and more general. 

In a highly differentiated society such as our own, Parsons 
argued, economic institutions are developed to deal primarily with 
adaptive problems. The institutions of defense and, in part, of 
education are also seen as falling into this sector. All are concerned 
with the provision of facilities having generalized means value in 
terms of social goals. When the system of reference is society as 
a whole, goal attainment clearly falls within the province of the 
state or polity. The integrative functions would seem to be divided 
among the state, the church, and many other structures about which 
important cultural values are focused. And what of the family? 
From the point of view of society as the reference system, the con- 
tributions of the family appear to fall mainly in the "latency sector"; 
the family, that is, speciahzes in the functions of pattern mainte- 
nance and tension management. When you have nowhere else to 
go and nothing else to do, you go home and the family will help 
get you ready for the next round. To this reviewer, it appears that 
the fit between system problems and institutions, as here depicted, 
is far from perfect. But even so, as with so many other Parsonian 
formulations, it has cast a new and provocative light on a number 
of familiar phenomena. 

In all this, so far, I have been treating society as a whole as 
the reference system, and institutional sub-systems as the component 
units. Let it be clear that each of these, whatever its own special- 
ized functions for society as a whole, also must face all four system 
problems. The business firm produces an output for society in the 
form of adaptive facilities; but within itself, it has its own problems 
of adaptation, goal attainment,, integration, and so on. These func- 

62 • Edward C. Devereux, Jr. 

tional needs may in turn be assigned to specialized departmental 
sub-systems, which also have to meet all four system problems in 
their own right. The result of this method of looking at matters is 
a long series of systems nesting within systems nesting within sys- 
tems, like a set of Russian Easter eggs. Needless to say, this poses 
some rather tricky problems for the analyst in keeping straight 
exactly which level of system-referent is being talked about. 

In Economy and Society Parsons exploits this nesting pattern to 
considerable advantage. The logic of his procedure is essentially 
this: within any given level of system, ignore what goes on inside 
its component sub-systems and attend only to what passes about 
among them, in the form of inputs and outputs. But don't stop here. 
After you have analyzed the exchange of inputs and outputs across 
sub-system lines, you must look inside the sub-system components 
and analyze them in the same way, looking for the sub-sub-systems 
which handle their own systems problems. Trace through what 
happens to the inputs received, how they are processed within the 
sub-system unit, and how the relevant output is generated. Note 
that what is goal attainment for the sub-system is delivered as out- 
put to the superordinate system. 

Keep this up for a rather long time, proceeding upward until 
you have reached society as a whole as your final system-referent, 
and downward until you are dealing with the exchanges within 
individuals between the organic system and the personality system. 
When you are done, you presumably have a complete account of 
the total action system of the society in question. And this account, 
be it noted, is not in terms of any static structural concepts but in 
terms of the continuous flow of phased and interlocking processes 
of interchange within and between systems. 

Parsons, of course, does not seriously recommend that anyone 
should actually try to do this in full detail, even if it were tech- 
nically possible. No more would an oceanographer attempt to chart 
and explain the exact positions and movement of every single wave. 
But in principle, with an adequate general theory, it ought to be 
possible. Fortunately for the sociological analyst, not all things 
are equally important. What Mr. Jones said to his wife at break- 
fast may have some importance for him and his own family, but 
by itself is not likely to have much effect on the institution of the 


family or upon society as a whole. What was decided at the 
bargaining table regarding wages and prices in the steel industry, 
however, may indeed have reverberations throughout the entire 
society. In his most recent discussions of this problem, Parsons has 
been talking in terms of a hierarchy of levels of organization and 
control. He has argued, for example, that within a business firm 
it is nonsense to consider the managerial function as simply coordi- 
nate with any other technical function. It is in a qualitatively 
diflFerent position because of its responsibility for determining the 
categories of inputs and outputs. But the fiduciary board of trustees 
occupies still a higher level in the hierarchy because of its concern 
with legitimizing the functions of the firm in terms of the values 
of the broader society.* 

Looking across the entire range of systems encompassed in his gen- 
eral theory of action, Parsons concludes that there is indeed an order 
among them: psychological systems organize and control the organic 
systems, social systems organize and control the psychological sys- 
tems, and cultural systems organize and control the social systems. f 
Though many other things have developed and changed in Parsonian 
theory over the years, institutionalized normative patterns still com- 
mand a major share of his interest and attention. 

* Talcott Parsons, "General Theory in Sociology," in R. K. Merton 
et al, eds., Sociology Today: Problems and Prospects (New York: 
Basic Books, Inc., 1959), ch. 1. 
\ See Koch, ed., op. cit., p. 616. 



Robin M. Williams, Jr. 









Professor Devereux has given us a summary and 
evaluation o£ an exceedingly complex body of thought set 
forth in a large number of publications over a period of 
more than twenty years. In this chapter I continue the 
analysis by further exposition and criticism of certain 
aspects of Parsons' works. The concepts and substantive 
problems touched upon at one time or another in the writings 
under review range over a large proportion of the major 
concerns of sociology and extend into anthropology, psy- 
chiatry, psychology, political science, and economics. This 
comprehensiveness is intentional; as early as the formu- 


lation of The Structure of Social Action in the 1930's, the central aim 
of Parsons' work was nothing less than the development of a con- 
ceptual scheme capable of subsuming all analytical knowledge of 
social conduct at a certain level of abstraction. The task of discerning 
and stating in a concise way the major elements in this ambitious 
program is not rendered easier by the circumstance that we have to 
deal not with the elaboration of a fixed set of concerns and ideas, but 
with a developing corpus of thought which has gone through a con- 
tinuous process of reformulation, often in quite subtle ways. 

The element of search and reformulation to which I have just 
alluded may be illustrated in several striking instances. The pre- 
occupation of The Structure of Social Action ( 1937 ) with the "unit 
act" contrasts with the quite small part played by this concept 
by the time of The Social System (1951); in the latter work the 
focus of attention has shifted toward "higher order" units such as 
"status-role" and "institution." Similarly, one finds some foreshadow- 
ing in the first of these two works of the later concern with psycho- 
analytic concepts, but Freud is mentioned just twice in the Index 
and the influence of his ideas is as yet obviously slight. In the later 
book, however, the Freudian themes have become strong and per- 
vasive. In the earlier work a concern with personality structures 
and processes is conspicuous by its absence; in the later formula- 
tions, personality not only takes its place alongside society and 
culture as one of the great "systems" within the theory of action, but 
also receives an impressive amount of attention even in a work 
explicitly focused on the social system. In reading the first major 
work one would hardly suspect that the socialization of children 
is a major process in social systems, but in Family, Socialization, 
and Interaction Process the topic has become a central preoccupa- 
tion. In 1937, the human organism is, for the most part, an inscru- 
table "black box" which in some unspecified way simply supplies 
the "energy" necessary to get "action" under way, and is of little 
further theoretical interest, except as an entity which is not the 
social actor. By 1951, considerable attention is devoted to such 
matters as the human infant's plasticity, dependency, affectivity, 
and capacity for symbol-mediated learning. A major interest of 
The Structure of Social Action is in the analysis of "rational action" 
and the differentiation of logical and nonlogical action; in the later 

66 • Robin M. Williams, Jr. 

works the concern survives, for the most part, only in the altered 
form of the attention given to the instrumental versus expressive 
behavior, and the lines of distinction have been considerably altered. 

All this is not to imply any fundamental discontinuity. On the 
contrary, one of the more impressive characteristics of Parsons' 
thought is its persistent, almost dogged, wrestling with a continu- 
ing set of really major theoretical problems throughout a long 
period of time. From the very beginning, the concept of institution 
has occupied a central position. Over and over again, now from 
one perspective, now from another, we are brought back to the 
broad question of the conditions for system-maintenance or "equi- 
librium." The part played by common values in social stability and 
change has been a focus of analysis in all of Parsons' major works. 
Continually, we find attention directed to the Hobbesian problem: 
how is order in society possible? But the question does not remain 
fixed and inert: its scope widens even as it becomes more complex 
and specific. In 1937 it is enough to say that social order is always 
a normative phenomenon; that the only ultimate preventive for 
the war of each against all is an agreement upon common values 
and symbols, reinforced by ritual; that neither coercive means nor 
advantaged interdependence nor both in combination are enough 
to account for the convergent mutuality of conduct that observa- 
tion finds to be so prevalent in known societies scattered through 
time and space. By 1951, the original assertions remain; but we 
now find that the existence of a common-value system has itself 
become a problem; it is no longer postulated simply as a logically 
derived formal condition for the existence of "orderly society." New 
explicitness characterizes the analysis of deviance, alienation and 
social control. Along with a continuing awareness of the massive 
societal import of power, political processes, and the economy of 
instrumental systems of action, there is in the later works a strik- 
ingly enhanced focus upon the micro-sociology of interpersonal 
relations. Consistent with this development is the elaboration of the 
analysis of expressive behavior and of systems of beliefs. 

In writings having to do with theoretical interpretations of human 
behavior one often senses a hidden dialogue in which we primarily 
hear one end of the conversation, and it is often helpful to know 
something more than the writer explicitly tells us about the views 


of the parties with whom he agrees or argues. As Devereux has 
noted, one can discern several polemical interests in Parsons' thought 
which have helped to shape the positive theoretical formulations 
which constitute the main body of his contributions. The targets 
of criticism emerge with particular clarity in The Structure of 
Social Action, a book best understood in the context of the prevail- 
ing intellectual currents in the social sciences during a period of, 
say, some twenty years prior to its publication. In that work Par- 
sons launches attacks across a wide front. He rejects, in the first 
place, biologistic and "instinct" theories as wholly inadequate to 
account for cultural variability and for the complex specificity of 
social conduct. At the same time he is equally decisive in pointing 
out the empirical and logical inadequacies of the various mono- 
factorial "determinisms": geographic determinism, economic deter- 
minism, and the like. On the opposite flank, he gives a trenchant 
critique of "radical rationalistic" formulations which conceive of 
action as determined by an actor's cognitively correct apprehen- 
sion of an environment— in which the rationally calculating actor 
is a sensitive high-speed scanning device, as it were, equipped 
with large-capacity electronic computing facilities. Against the 
vogue of neobehaviorism and elementaristic stimulus-response in- 
terpretations, he contends that much behavior is goal-directed, 
that "ends" are not epiphenomenal, that action is normatively de- 
fined and regulated, that values do exist and have an independent 
causal efficacy. Against conceptions of society as a "symbiotic" or- 
der of economic and political interdependence, he argues that via- 
ble societies cannot exist without a minimal sharing of values, going 
beyond considerations of sheer expediency. To an American sociol- 
ogy wedded to empiricism on the one hand, and bemused by prag- 
matic ameliorism, on the other, he brought a new insistence upon 
the legitimacy, necessity and fruitfulness of systematic abstract 
theory, together v^ith the astringent perspective on "social prob- 
lems" of such European scholars as Max Weber and Emile Durk- 

* The substantial impact of these views upon sociology in the United 
States was perhaps less diminished than accentuated by the fact that 
the book in which they appeared was learned, long, complex, ab- 
struse, and difficult. The challenge was impressive, and in manv 
quarters it soon became a mark of prestige to have "read Parsons. 

68 • Robin M. Williams, Jr. 


Although the full exposition of the developed body of the 
"theory of action" is elaborate and, in a certain abstract way, de- 
tailed, the initial set of conceptions with which the schema begins 
can be stated rather briefly. 

We start with an actor ( ego ) in a situation. The actor is a more- 
or-less socialized human being, endowed with the organismic 
characteristics of energy, capacity for learning, dependency and 
vulnerability, high sensitivity to stimuli and the ability to discrimi- 
nate and generalize among them, bi-sexuality, capacity to use sym- 
bols and to remember and to anticipate, mortality, and a number 
of other significant properties that are "normally" elicited in the 
life-course of an individual living in society. The situation within 
which the actor acts is composed of physical objects (including 
organisms other than men), cultural objects, (artifacts, language, 
value-and-belief systems, symbols of various kinds— insofar as these 
are not directly constitutive of social actions * ) and social objects, 
which may be either individual social actors or collectivities. "Ac- 
tion" is behavior that is, in some sense, directed toward goals; it 
has motivational significance; it is not just a specific reaction to a 
momentary stimulus, but rather has some systematic quality, es- 
pecially insofar as it involves "expectations" as to the contingent 
actions of other social actors. 

In a specific situation a specific actor's motivations may be re- 
garded as classifiable into cognitive, cathectic,t and evaluative 
modes. He must identify objects and define their characteristics 
relevant to his interests, appraise their gratificational possibilities, 
and— at least once we go beyond a single elementary act— evaluate 
alternative cognitive interpretations and different possibilities of 
cathectic gratification. 

Now, "action" is defined as behavior in which these several as- 

* ", . . are treated as situational objects by ego and are not 'internal- 
ized' as constitutive elements of the structure of his personality." 

f "Cathexis, the attachment to objects which are gratifying and re- 
jection of those which are noxious. . . ." (GTA 5) 


pects of objects are interpreted in terms of shared (cultural) sym- 
bols. Indeed the intent in The Social System is to restrict analysis 
to ". . . systems of interaction of a plurality of individual actors 
oriented to a situation and where the system includes a commonly 
understood system of cultural symbols." (SS 5) Given such a 
symbolic system, then, the possibility arises of standards or criteria 
by which "selection" is made among the various orientations 
possible in a situation; these standards are called values ( or "value- 
orientation aspects" of action). Corresponding to the three motiva- 
tional modes, there are cognitive, appreciative, and moral standards. 
"Moral" values, in this sense, are standards for judging the syn- 
thesis of cognitive, cathectic, and evaluative motivations, together 
with cognitive and appreciative standards, that issues in concrete 
action. Motivations are logically independent of values, e.g., know- 
ing an actor's cathectic motivation does not permit us to deduce 
the appreciative standards he will apply in appraising the cathected 

Motivated actors seeking gratifications and oriented to shared 
values or standards thus interact in patterned ways. The total ac- 
tion system thereby constituted may be thought of as composed of 
three interpenetrating and overlapping, but conceptually distinct, 
sub-systems: culture, personality, and social systems.* 

At the time of The Social System and Working Papers, the cul- 
tural system consists of the entire "social heredity" of shared prod- 
ucts of social activity— language, ideas, beliefs, values, art, law, 
etc.— insofar as these are objects of orientation but not constitutive 
of personalities and social interaction patterns. "A cultural system 
does not function except as part of a concrete action system, it 
just 'is.'" (SS 17) Culture, then, is a part of an action system only 
to the extent that it is "internalized" in personalities or actually 
defines appropriate interaction. Otherwise, it is inert— presumably 
in the sense that cultural items are transferable from one society to 
another, put in museums, or treated in a detached way as objects 
of intellectual interest, as, for example, Mayan art or ancient Egyp- 
tian marriage ceremonies. The cultural system is integrated, to the 

* A fourth system, the organism, was brought into the scheme at a 
later date; see p. 70 below. 


70 • Robin M. Williams, Jr. 

extent that it is, in terms of pattern-consistency such as logical 
coherence or aesthetic "style." The question of how a culture which 
does not "act" can be part of a system of "action" turned out to be 
somewhat troublesome. It is possible to say that culture "enters 
into" action when it becomes "constitutive of" a system of social 
interactions, but one is immediately led to wonder what this means 
and how one could possibly disentangle the cultural from the so- 
cial in observed behavior.* More recently Parsons has proposed 
that the study of culture as a timeless, pure, "symbol-meaning sys- 
tem" belongs to such formal disciplines as logic, aesthetics, and 
ethics. t What then remains as part of the theory of action is that 
part of culture which has to do with the creation and maintenance 
of symbol-meaning systems. The cultural system, in this new for- 
mulation, thus becomes a sub-system of a total concrete system 
of action. 

The personality system has its focus in the motivational inte- 
gration of the socialized human individual. The personality system 
is a property of a single living organism. J Although social relation- 
ships are directly a part of this system, the functional problems 
diflFer from those of the social system. A personality must "come 
to terms" with the demands and expectations of other persons in 
relation to its own unique organismic and other socially idiosyn- 
cratic needs. It is, in one aspect, a bundle of "need-dispositions," 
and the focus of integration lies in the balancing of these needs 
against one another, both momentarily and through time. A social 
system, on the other hand, is a network of interactions, and cannot 
be predicted from a knowledge of individual personalities taken 
one by one. Its focus of integration is the balancing of interactive 

* See the concise critique by Marion J. Levy, Jr., "Some Questions 
About 'The Concepts of Culture and of Social System,' " American 
Sociological Review, 24, No. 2 (1959), pp. 247-48. 
f "A Rejoinder to Ogles and Levy," American Sociological Review, 
24, No. 2 (1959), pp. 249-50. 

I Originally Parsons worked with the three systems of society, cul- 
ture, and personality. Later, a fourth system, the organism, was 
added. A psychological system is a system of action characterized by 
the fact that all the behavior belonging to it is behavior of the same 
living organism. ["An Approach to Psychological Theory in Terms 
of the Theory of Action," in Systematic Theories in Psychology, Sig- 
mund Koch, ed. (New York: McGraw-Hill, 1960), pp. 612 ff.] 


sets— a process which to be "successful" must allow for biological 
and personality needs, for cultural integration, and for adaptation 
to the physical world and the external social environment. 

Contrary to the interpretation of some critics of this scheme, we 
note that Parsons repeatedly and emphatically calls attention to 
"strains" and "inconsistencies" within each of these four analytically 
separable systems, and among them. The fundamental image dis- 
cernable behind the conceptual scheme is that of an energized 
network of interactions among goal-seeking personalities, whose 
goals and concrete motivations are partly shaped by a shared set of 
norms and symbols, and who must cope with survival-problems 
vis-a-vis the physical environment and the actions of other so- 
cieties and collectivities. Inherent in social action are tendencies 
toward deviance and alienation of individuals, toward inconsist- 
encies in cultural patterning, and toward secession, schism, and 
conflict in the relations of sub-units of social structure. 

It remains true that the main preoccupation of the body of 
theory under examination is to account for order, stability, and 
equilibrium rather than for disruptive or violent change. The basic 
model from which analysis departs is that of a boundary-maintain- 
ing system in which small changes are counteracted in such a way 
as to restore the prior state of aflFairs, or else to produce "orderly" 
change (presumably, gradual and nonviolent). In the microscopic 
case of ego-alter relations, the point of departure is a relation in 
which there is exact mutual comprehension of meanings, full com- 
plementarity of expectations, consensus upon standards of evalua- 
tions, and optimal mutual gratification. It must be said that such 
relations do find approximate empirical embodiments, as in Aris- 
totle's "true friendship," in certain cases of romantic love, in the 
serenity of spouses who have experienced a long and happy mar- 
riage—or in the exultant mutuality of strong and well-matched op- 
ponents in games or foes in combat. In the perfectly integrated 
relationship, complete reciprocal conformity is induced in four 
ways: (1) ego and alter act in such manner as to gratify directly 
specific need-dispositions of the other; (2) the actions of each are 
instrumentally useful to the other in the attainment of still other 
goals; (3) because of the internalization of shared values, con- 
formity to the legitimate expectations and demands of the other 


72 • Robin M. Williams, Jr. 

person is directly gratifying in its own right; (4) because of so- 
cialized sensitivity to the attitudes of others toward oneself, the 
approval and esteem elicited from the other person through "con- 
forming" behavior is directly rewarding. 

Interaction involves a "double contingency": the satisfaction of 
ego's needs is contingent upon the actions of alter, but alter's ac- 
tions are to a large extent shaped by the actions of ego in the first 
place. For there to be dependability in the two-way flow of gratifi- 
cations, each party must be able, to some important degree, to 
"predict" (anticipate) the relevant actions of the other. It is in 
this connection that common values play a crucial part: predicta- 
bility is facilitated by shared commitment to cognitive, appreciative, 
and moral standards. These "norms" help to specify the goals to be 
sought, the means to be employed, and the permissible direct 
gratifications to be obtained within the immediate relationship. 

Thus, at the deepest levels of personality and in the most ele- 
mentary forms of interaction. Parsons seeks to resolve the Hobbesian 
problem of order in society by postulating interlocking "interests" 
in need-gratification which are, at the same time partly defined by, 
and integrated with, mutually held criteria of evaluation. 

A word must be said at this point concerning two aspects of 
"moral" values which appear in this argument. The first refers to 
those evaluative standards in terms of which the personality sys- 
tem is integrated ( "ego-integrative" moral values), that is, the 
standards utilized to choose among various cognitive and cathectic 
possibilities to sustain an optimal flow of gratifications through 
time. But each personality is deeply involved with others and its 
ego-integrative "solutions" must, in the social system, confront the 
problem of compatibility with the solutions attempted by alters 
with whom ego interacts. Hence the personality-integrative moral 
values are not sufficient to cope with the realistic exigencies of a 
functioning social system. This second aspect of the problem has 
been summarized by William L. Kolb. 

Although certain areas of such a social system can be integrated 
through orientation to instrumental standards and appreciative 
standards governing shared orientations toward means and ends 
on the one hand, and ordered series of gratifications on the 
other, tlie over-all integration of the system can be obtained only 


by mutual orientation toward shared moral value-standards. 
These values perform the function for the social system that 
personality-integrative values perfoiTn for personality. They de- 
fine a mode of social-system integration both as an ideal, and 
at the actual organizational level, as a sanctioned achievable 
end. Further, they define ideal and expected rights and obliga- 
tions of the actors in their direct relations with one another, and 
they control and limit the range of private ends and the means 
used to achieve them insofar as such means and ends impinge 
on the integration of the social system.* 

As soon as we turn from the elementary social act to a considera- 
tion of the social system as such we confront the problem of choos- 
ing a "unit" with which to begin analysis. Parsons says: ". . . for 
most purposes of the more macroscopic analysis of social systems 
... it is convenient to make use of a higher order unit than the 
act, namely the status-role as it will here be called. Since a social 
system is a system of processes of interaction between actors, it is 
the structure of the relations between the actors . . . which is es- 
sentially the structure of the social system. The system is a network 
of such relationships." (SS 25) The status is the location of the 
actor; the role is what he does in that position, in aspects significant 
for the system of relationship. Status consists of the rights and obli- 
gations of ego seen as a social object by others; role consists of ego 
performance as subject in reacting to the actions directed toward 
him as an "occupant" of a status. Role is defined in terms of norma- 
tive expectations: it is ". . . integrated with a particular set of 
value standards which govern interaction with one or more alters 
in the appropriate complementary roles." (SS 38-39) Since each 
person occupies a number of different statuses, the organized sys- 
tem of statuses and roles referable to him as an individual consti- 
tutes the social actor. Systems of statuses abstracted from social 
actors may be combined into collectivities, which are partial so- 
cial systems to which actors have a definite obligation of solidarity 
("responsibility" for maintaining and defending the aggregate). 
( SS 41, 97-101 ) A complex social system ". . . is to be regarded as a 
network of collectivities side by side, overlapping and larger- 

* "The Changing Prominence of Values," in Howard Becker and 
Alvin Boskoff, eds.. Modern Sociological Theory (New York: The 
Dryden Press, 1957), p. 116. 

74 • Robin M. Williams, Jr. 

smaller." ( SS 101 ) Each collectivity is a system of concretely inter- 
active specific roles. 

Before going on to show further how single roles are built into 
networks and sub-systems, we must return to the step-by-step con- 
struction of interrelated concepts, so characteristic of the whole 
body of work under examination. 

From the perspective of the maintenance or change of a social 
system, reciprocal role-behavior may be regarded as a pattern of 
social control. This is true not only because the acts of a role- 
partner may be useful or disadvantaging to us in terms of the pos- 
sibility of attaining specific goals which we desire apart from the 
immediate content of the relationship. It is true also in a more 
direct way: just to the degree that we have developed "sensitivity" 
to the attitudes of others, their approval or disapproval of our spe- 
cific behavior, and especially of us as total personalities, will be 
rewarding or punishing— and this quite apart from whether their 
acts toward us were intended to have this significance. It is in this 
light that any role-behavior may be regarded as a sanction. Sanc- 
tions, in fact, are defined as role-expectations seen in terms of their 
gratificational significance, as rewards or punishments. (SS 38) * 

Given that there are statuses occupied by actors holding re- 
ciprocal role-expectations, involving shared value-standards, we 
are quickly led to the still more complex notion of institution which 
is ". . . made up of a plurality of interdependent role-patterns or 
components of them." An institution is a complex of institutionalized 
"role-integrates" (or "status-relationships") ". . . which is of stra- 
tegic structural significance in the social system in question." (SS 
39) An institution is thus a pattern of rights and obligations or- 
ganized around some functional focus. On the other hand, a col- 
lectivity is a system of concretely interactive roles, involving senti- 

* In passing, we should note the puzzling difficulty of determining 
when Parsons is speaking of action as action and when he is referring 
to "psychic states" or "attitudes." The term "role-expectation" seems 
particularly difficult to fix firmly in place as meaning either "expecta- 
tion" in the sense of a subjective disposition of an actor, or else an 
"act of communication" wliich conveys to alter that ego has a 
certain expectation. It is only slightly less difficult to keep in mind 
that "expectation" does not mean only passive anticipation but also 
includes some quality of active demand. (Cf. SS 5-7.) 


ments of solidarity and sanctioned obligations of responsibility on 
the part of the actor who is a member of that particular system 
of interaction. A single institution may appear in many different 
collectivities, and several different institutions may be found in 
one and the same collectivity. 

Institutionalization means that action is being guided by shared 
"moral" values that have been "internalized" by social actors in 
such a way as to become "genuine need-dispositions of the per- 
sonality." (SS 42) In the theoretical perfect case, conformitv to 
institutionalized role-expectations brings gratifying responses from 
alters, is instrumentally effective, and is a source of direct gratifica- 
tion as well. Everyone wants to do that which others want him to 
do, and others always act as he expects and wishes: ". . . the 
interests of the actors . . . [are] bound to conformity with a shared 
system of value-orientation standards." (SS 38) Although such 
perfect integration is a limiting case, not found empiricallv, this 
mode of normative integration is to be regarded as fundamental 
in all actual social systems. This tying-together of need-dispositions 
with values is the point of reference for what Parsons calls the 
"sociologistic theorem." "This integration of a set of common value 
patterns with the internalized need-disposition structure of the 
constituent personalities is the core phenomenon of the dynamics 
of social systems. That the stability of any social system except 
the most evanescent interaction process is dependent on a degree 
of such integration may be said to be the fundajnental dynamic 
theorem of sociology." (SS 42) 

Now, given the fact of institutionalization, we proceed to classify 
institutions. There are, first of all, the basic relational Institutions, 
i.e., those which directly define the statuses and roles in the net- 
work of interactive relationships. Within relationships so defined, 
however, a further problem of normative order arises because 
actors in their pursuit of instrumental, expressive, and ego-inte- 
grative (evaluative) interests may act in ways that are disruptive 
for the functioning of the system of interaction. Here arise regu- 
lative institutions which limit the goals sought and the means 
employed. Third, cultural institutions are those which concern the 
sheer acceptance of patterns of cultural orientation, without com- 
mitment to overt action. Presumably the role-patterns in this case 


76 • Robin M. Williams, Jr. 

consist merely of the acceptance of beliefs, expressive symbols, or 
moral values in a manner which does not involve any norms for 
interaction beyond such acceptance. That it may turn out to be 
operationally di£Scult to distinguish "sheer acceptance" from "com- 
mitment for action" is suggested by Parsons' admission that ac- 
ceptance may lead to commitment, as when ". . . subscription to a 
system of belief becomes a criterion of loyalty to a collectivity, 
such as a religious group." (SS 56) 

At this point, then, we have a classification of motivations, of 
value-orientations, of culture patterns, of interests, of evaluative 
action-orientations (instrumental, expressive, and moral), and of 
institutions. Parsons now asks how we may analyze the basic al- 
ternatives of an actor in defining his relationship to an alter, and 
chooses to focus upon ". . . the coUectivity-integrative sub-type of 
the moral type of evaluative action-orientation." (SS 59) The an- 
swer to the question consists of the five "pattern-variables"* already 
well described in Devereux's discussion {supra, p. 38). As noted 
in that chapter, the pattern-variables have been widely utilized 
in sociological codifications, interpretations, and first hand research. 
It remains to be demonstrated that the listing is "exhaustive," 
even at its chosen level of generality, and the exact denotation of 
the terms remains to be fully established. It is not an easy task to 
translate the concepts into specific indicators, and it may well be 
that the concepts will have to be modified in their adaptations to 
research utilization. 

In a recent eflFort to give operational meaning to the concept 
of specificity-diffuseness in a study of friendship patterns in a sub- 
urban community, it appeared that the pattern-variable refers to 
two partly separable aspects of the norms governing an interper- 
sonal relationship. Taking the description given in Toward a Gen- 
eral Theory of Action, for example, we find that diffuseness is rep- 
resented by the role-expectation that the actor 

. . . will accept any potential significance of a social object, in- 
cluding obligation to it, which is compatible with his other 

* "Pattern-variables" because the reference is to normative patterns 
each of which varies along a continuum from one polar opposite to thr 



interests and obligations, and that he will give priority to this 
expectation over any disposition to confine the role-orientation 
to a specific range of significance of the object. On the other 
side, specificity is defined by an expectation that the actor will 
be oriented to the social object only within a specific range of 
significance and will give priority to this orientation as over 
against any disposition to include other aspects of significance 
not already specifically defined in the expectation pattern. 

It appears that specificity has two aspects: (1) whether ex- 
pectations as to rights and obligations are highly restiicted or 
relatively unlimited; (2) whether the rights and obligations are 
clearly defined or not. Note the following statement: 

The rights of a social object with respect to ego are either defined 
(so that ego and alter know the limits of ego's obligations) or they 
are undefined (so that ego must render to alter much of his efforts 
as are left over when all of his other obligations are met). The 
social object, that is, either has specific (segmental) significance for 
ego (in which case obligations are clearly defined); or it has diffuse 
significance (in which case obligations are only limited by other 
obligations ) . 

In this case we are forced to make a further distinction within 
the definition of the pattern-variable. As we began to construct 
interview questions aimed at the diffuseness-specificity aspect of 
friendship, we found that some items seemed to index width of 
range of significance, whereas others dealt with the explicitness 
of definition of rights and obligations. For example, respondents 
were asked whether in their relationship with their one best 
friend in the residential area, the friend should feel free to dis- 
cuss intimate personal matters. The question asks about range 
of intimate topics. 

of Normative 

Range of significance of social object 







Example: "We discuss 
anything except reli- 
gion and politics." 



Example: Loan money 
without specifying re- 


On the other hand, a question could deal with a narrow area 
of interests and direct itself to the explicitness of obligations. 

78 • Robin M. Williams, Jr. 

e.g., a question concerning the functionally specific action of 
loaning money when the debtor is defined as a friend. . . .* 

The pattern-variables become a major basis for classifying social 
systems as wholes and for analyzing social roles (e.g., the medical 
doctor) in The Social System, and later for characterizing phases 
of action in group processes, in the Working Papers, and in Fam- 
ily, Socialization and Interaction Process. Of all the components 
of the Parsonian scheme they have been most often noted and 
used by other social scientists. The categories have developed in 
Parsons' thinking over an extended period; universalism and par- 
ticularism, for example, were already prominent at the time of 
The Structure of Social Action (cf. pp. 547-551), having been 
drawn originally from Max Weber's analyses of world religions 
from F. Toennies' Gemeinschaft und Gesellschaft (cf. sa. 686-694). 
Having been formulated partly out of an interest in the compara- 
tive study of institutions,! the pattern-variables are especially 
likely in Parsons' work to be turned to comparisons of social struc- 
tures. At the same time, however, they are extensively employed 
to characterize specific role-structures. 

With the formulations just reviewed, we may turn to focus upon 
relational institutions— regarded as the core of the social system— 
in order to show how types of action-orientations, roles, institutions, 
and constitutive values are combined in sub-systems of societies. 

Although a rather complex set of distinctions have been re- 
viewed thus far, the conceptual scheme is due to unfold much 
further. The process of development is that of setting forth suc- 
cessively closer approximations to concrete social structures and 

In the first place, given the universality of kinship units and the 
space-bound limitations of social living, there necessarily arise 
ascriptive and diffuse territorial groupings called communities. 
Assuming the emergence of distinctive cultural characterictics 

* R. M. Williams, Jr. et al.. Friendship and Social Values in a Sub- 
urban Community (University of Oregon, 1956), pp. 30-32. 
f In a course offered by Parsons on "Comparative Social Institutions" 
in the late 1930's, most of the current distinctions were already in 
use, plus a rationalism-traditionalism polarity, later dropped out of 
the set. 


shared among a number of kinship units (e.g., language or reli- 
gion), "intergroup" contact leads to a "horizontal" aggregating of 
these units into ethnic collectivities. At the same time we will find 
that kinship units are not all valued or ranked identically; sys- 
tematic differences in prestige arise and a rank-order of social 
classes emerges; each "class" is an aggregate of kinship units of 
roughly equivalent rank. These four structural components will be 
found in every self-subsistent society. 

Social differentation may occur also in the direction of the segre- 
gation of specialized, functionally specific roles out of the diffuse 
matrix of kinship and communities. A very important development 
of this kind consists of the elaboration of chains of instrumental 
acts involving the transfer of valued objects and services in trans- 
actions of "mutual advantage." In this "economy" of transfers, as 
roles become more specialized and numerous, a true exchange 
system can arise, in which direct reciprocity within a solidary col- 
lectivity gives way to a situation in which ego may be remunerated 
from one source, have his produce used by another part}% secure 
productive facilities from a third, and cooperate with still others 
in the productive process— all these relationships being divorced to 
an important extent from purely ascriptive rights and obligations. 
The existence of such a circular flow of relatively "free" transfers 
necessitates (or, at least, creates a pressure toward) some com- 
monly accepted ground rules governing access to facilities, control 
of possessions, terms of settlement of exchange, and differential 
rewards (remuneration). Broadly speaking, these rules constitute 
"economic institutions." 

A second major line of differentiation concerns the acquisition 
and exercise of power in society, particularly of coercion and 
physical force within a particular territory. More generally, there 
are the functional problems of resolving conflicts, compromising 
differing interests, and integrating divergent values within a social 
system made up of differentiated units. The "control of power," in 
one major aspect, is the area of political institutionalization. \\^hereas 
economic "power" is additive (a matter of having more and more 
units of control over facilities, possessions, and remuneration), polit- 
ical power is inherently hierarchical; it is a matter of power over 
lesser power. Parallel to the relational system involved in instru- 

80 • Robin M. Williams, Jr. 

mental actions is a set of institutionalized rights in "relational 
possession," i.e., a system of regulated expressive actions. Corre- 
sponding, by analogy, to "facilities" is an appropriate context for 
expression; in place of "disposal of product" in the instrumental 
case, we now have "receptiveness" of alters; for "remuneration" we 
read "responsiveness"; "cooperation" is replaced by "expressive 

Even with established instrumental and expressive systems, in- 
stitutionalized in a network of roles, there remains a problem 
". . . of establishing the patterns of order both within the instru- 
mental and the expressive complexes respectively, and between 
them, since every actor must have relationships of both types." 
(SS 80) Ego-integrative solutions for each actor's unique person- 
ality and relational systems are not enough to account for moral 
integration or order in the social system as such. 

This problem is approached by a complex classification of pos- 
sible combinations of the pattern-variables, as related to the fusion 
or segregation of instrumental and expressive interests, and fol- 
lowed by a description of "the modalities of objects as foci of role- 
expectation." The latter refers simply to the major characteristics 
of alters by which they are defined and classified in interaction, 
e.g., sex, age, physical traits, territorial location, status classes or 
categories, membership in a collectivity. Achievement criteria of 
object-selection, on the other hand, refer to actual expected, spe- 
cific performance. Achievement is necessarily judged by universal- 
istic standards, although of course there can be successful per- 
formance in the service of particularistic values; achievement for a 
collectivity is tied to social relations, but given the goal, success 
may be judged in universalistic terms. 

Having thus reviewed (1) types of orientation of actor as ego 
(e.g., combinations of neutrality, universalism, specificity), and 
(2) orientations (in terms of ascription-achievement) to social ob- 
jects, we must as our next step see what implications may be drawn 
from self vs. collectivity orientations. The starting point is the 
observation that there can be a sharing of cognitive standards of 
communication or appreciative standards governing expressive 
symbols without a sufficient sharing of moral orientations to con- 
stitute a collectivity. The important point is whether there is agree- 


ment of the members on what actions are "required" in the interests 
of keeping the system going (e.g. accepting mihtary conscription, 
paying taxes, spending time in faculty committees). Conformity to 
collectivity-demands takes two forms : ( 1 ) loyalty, which is a "spill- 
ing over" of motivation to conform, beyond institutionalized obli- 
gations, and (2) soUdarity, which is an obHgation; it is demanded 
and sanctioned, whether one "spills over" or not. 

Collectivities may be classified according to whether they are 
predominantly characterized by primacy of expressive or of instru- 
mental interests: the familiar dichotomy of Gemeinschaft vs. Cxesell- 
schaft. Collectivities may also be classified according to their modal 
or typical combinations of primacies of the pattern-variables. 

The main functional problems of any social system are those of 
integration and allocation (adaptation). The ways in which these 
problems become focal points of institutionalization may be shown 
by returning to the "economic" and "political" sectors already 
briefly noted. Because a social system is made up of differentiated 
roles, there must be an allocation of roles and of persons among 
roles. Supposing this allocation to be accomplished, there is still 
the problem of allocating facilities and rewards, both of which 
are possessions, i.e., bundles of rights which are transferable be- 
tween actors. Given transferability and scarcity, an orderly ex- 
change economy depends upon at least three basic conditions: 1) 
the development of processes of settlement of terms in an extended 
system of differentiated roles; 2) high development of universal- 
istic norms; 3) institutional control of the most drastic means of 
exercising power. It follows that economic and political orders are 
inherently interdependent, although conceptually distinguishable. 
The extension of "economic" activities depends upon the insulation 
of exchange relationships from diffuse and particularistic structures 
and upon the limitation of force, fraud, and certain other disruptive 
factors. Although all control of productive facilities and of rewards 
is "power," this power is specific and strictly limited in scope. 
Political power, on the contrary, is generalized and diffuse: it is 
"capacity to control the relational system as a system," and oper- 
ates directly on specific relational-sets, arranged in hierarchical or 
concentric systems upon systems. The irreducible sub-stratum of 
political power is the ability to use force in relation to a territory. 

82 • Robin M. Williams, Jr. 

From a similar analysis of the relational reward system of ex- 
pressive action, Parsons identifies approval as specific and affec- 
tively neutral evaluation of a specific quality-complex or perform- 
ance, and esteem as diffuse and affectively neutral evaluation of 
actors. Like political power esteem tends toward hierarchical order- 
ing, hence constitutes a ranking system of stratification. It will be 
noted that by the distinctions just outlined Parsons has decisively 
moved "stratification" out of the political and economic orders. Al- 
though economic and political power may still affect stratification, 
the ranking system itself is conceived as an ordering in terms of 
prestige, a diffuse evaluative judgment of alters. 

The differentiated sub-systems now outlined are all simultane- 
ously operative within collectivities and networks of collectivities 
tied together into larger social systems. What are the integrative 
foci for these aggregative systems? There is the regulation of allo- 
cation of roles and their interrelations (and changes) of personnel, 
and of facilities, rewards, political power, and prestige. The actual 
institutionalization of regulation may be in the form of private, 
spontaneous sanctions, or of formalized sanctions (which require 
the development of "specialized" roles of responsibility). 

This completes the main outline of social system components and 
their interrelations as set forth in The Social System."^ Parsons 
maintains that filling in the categories thus defined ". . . with the 
requisite detail of properly conceptualized statements of empirical 
fact will constitute an adequate description of a concrete social 
system, the amount of detail required depending on the problem." 
(SS 138) 

Up to now, we have mainly seen only the structural part of the 
scheme, and that only in bare and abstract form. Some further 
insight into what is being attempted perhaps may be gained by 
a quick review of the application of the scheme to the description 
of some aspects of "internal differentiation and comparative vari- 
ability to types of social structure." ( SS 151 ) For reasons of brevity 
I shall present only a few illustrations of this application, in the 
form of an annotated outline. 

1. Kinship. Out of a very large number of possible combinations 

See SS 136-137, 142-159 for a convenient summary. 


of elements, only a very few are actually used in kinship systems. 
Child-care and status-ascription of infants always attach to kinship 
units; there is always an incest taboo; kinship roles are always 
diflFuse and collectivity oriented. The ubiquity and persistence of 
kinship units with these characteristics suggest that these particular 
clusterings result from powerful and universal forces. 

2. Instrumental achievement and stratification. With specializa- 
tion and achievement-emphasis, a high degree of division of labor 
results in a wide range of evaluated "competence." Instrumental 
role-diflFerentiation requires organization; organization leads to roles 
with diflFerent degrees of "responsibility." These diflferences "require" 
diflFerential access to and control of facilities. And this signifies 
differential rewards. Ergo: equality of reward is highly improbable 
in a complex division of labor. And because of the functional 
characteristics of kinship, differential advantages tend to be passed 
on to children. 

3. Territoriality, force, and the integration of the power system. 
For reasons already stated, power easily becomes the focus of dis- 
ruptive conflicts. Some regulation of drastic means of power is 
essential to the maintenance of a social system, although the kind 
and effectiveness of this control of power varies enormously. 

In a striking summary Parsons says: "We may conclude, then, 
that societies where there is almost unrestricted freedom to resort 
to force, and above all where several agencies with independent 
control of organized force operate within the same territorial area, 
are as rare as societies where children are socialized without any 
reference to kinship relations or where the reward system is in 
inverse relation to the graduations of competence and responsibility 
in the principal areas of valued achievement." (SS 163) 

4. Religion and value-integration. Religion may be conceived as, 
in part, one response to the problems of death, of imperfect control 
of the physical world, of malintegration of society (the agonizing 
impact of "undeserved suffering" and "unpunished behavior"). The 
beliefs and expressive symbols which are concerned wdth these 
problems must bear some definite relation to the "dominant" system 
of [nonreligious?] institutionalized values. Organized religion, 
especially, cannot wholly divorce itself from "secular" social struo 
tures and processes. 


84 • Robin M. Williams, Jr. 

Without going further into other illustrations, such as the analysis 
in terms of pattern-variables of principal types of social structures, 
it is necessary to turn to the place of process in the structure. A 
process is the way in which a change from one state of a system 
to another occurs; a mechanism is just a process looked at in terms 
of the significance of its effects upon the system or some part of it. 
For instance, what is learning in the personahty becomes socializa- 
tion of the person in the social system.* Personality defenses (deal- 
ing with conflicts of need-dispositions) and adjustments (dealing 
with strains and conflicts in relation to objects ) appear in the social 
system as problems of deviance and social control. For example: 
"A mechanism of social control, then, is a motivational process in 
one or more individual actors which tends to counteract a tendency 
to deviance from the fulfillment of role-expectations, in himself or 
in one or more alters." (SS 206) 

The basic learning processes are discrimination and generaliza- 
tion. Through these, there are built up five types of cathectic- 
evaluative mechanisms, viz.: (1) reinforcement-extinction (the 
strength of the tendency to repeat an action); (2) inhibition (to 
refrain from gratification); (3) substitution (transfer of cathexis); 
(4) imitation (taking over specific items from a model); (5) iden- 
tification ( internalizing the values of a model ) . Parallel to reinforce- 
ment-extinction are reward and punishment as social mechanisms; 
parallel to imitation is instruction; the social counterpart of identi- 
fication is attachment. 

Great importance is given, in this scheme, to early attachments, 
which are held to be necessarily specific and affective initially, grow- 
ing into diffuse relations, e.g., to the mother. The early diffuse attach- 
ments constitute the child's security system, but by the same token 
constitute marked dependency. How is dependency broken through? 
Speaking very generally, the answer is that the child within a 
"normal" diffuse love attachment develops a tolerance for frustra- 
tion which makes it possible for the socializing agents to guide 
him toward affective neutrality, universalism, achievement orienta- 
tion, and functional specificity. (SS 219) This movement, it is 
hypothesized, is favored by adequate security, imposition of disci-- 

Although not all learning is "sociaHzation/ 


plines, permissiveness for adjustive responses evoked by frustration, 
and use of rewards, especially relational rewards. It is further 
hypothesized that the order of difficulty in learning evaluative 
moral orientations is from affectivity to neutrality, from particu- 
laristic to universalistic, from aflFective specificity to difiFuseness. 
Further: "The orientation element, which is most difficult to acquire 
and which in a sense depends on the most complex set of pre- 
requisite conditions, is, at least under certain types of strain, likely 
to be the first to break down." (SS 226) In the analysis of social 
systems it is of great importance to know the specific modes of 
socialization and the strains they engender, as well as to identify 
the ways in which different social structures lead to different con- 
sequences in personality processes. , 

Furthermore, in socialization ". . . the combination of value- 
orientation patterns which is acquired must in a very important 
degree he a function of the fundamental role structure and domi- 
nant values of the social system." (SS 227) To the extent that this 
supposition is true, social systems will exhibit a modal basic per- 
sonality structure, which nevertheless is subject to considerable 
diversity. The variation or diversity means that we cannot infer 
directly from basic personality to social system. Rather we must 
go on to look for capacities of individuals for "rational adaptations" 
to varying situations, for additional mechanisms of socialization, 
and for mechanisms of social control. Parsons emphasizes the 
multiple and complex sources of variations in the social outcomes 
of socialization and personality mechanisms, the existence of alter- 
native role-opportunities and of ranges of tolerance for deviation. 
Furthermore, he attaches great importance to situational specificity 
in which the generalized need-dispositions and values of the 
personality are defined in detail, often differing markedly from 
what would have been predicted from "basic personality structure" 

These considerations lead to an extended discussion of deviance 
and mechanisms of social control. This far-ranging and often pene- 
trating and perceptive analysis is too extended for detailed review 
here. It is important to note, however, that throughout his dis- 
cussions of conformity and deviance, Parsons stresses the variability 
of individuals' responses to social demands, the continuous "veering 

86 • Robin M. Williams, Jr. 

off-course" of social actors from the cultural blueprint. He sees 
this recalcitrance to conformity arising from constitutional differ- 
ences among individuals, idiosyncratic, learning, unique status-role 
combinations, and from several other sources. It is possible, how- 
ever, that the very model used to depict conformity tends to deflect 
attention away from some of the important sources of mal- 
integration. For example, it has been suggested that relations of 
complementarity— in which ego's rights correspond to alter's duties 
or alter's rights correspond to ego's duties— may be subject to 
endemic strain because of the diflBculty of "equating" exchanges of 
gratifications and because of the high likelihood of "egoistic" tenden- 
cies on the part of the interacting parties.* Or, again, the basic 
ego-alter model in the Parsonian scheme does not directly take 
into account differences in power, in the sense of unequal capacities 
to control the relationship, which may and do make crucial differ- 
ences in the character of the interaction. f Still a third possible 
source of deviation from a non-problematic state of balanced com- 
plementarity would be of importance should it turn out that 
successive acts of conformity in a series of interactions have decreas- 
ing reward-value. In that case, clearly, the longer the conformity- 
series, the less the potency of each successive act of conformity by 
alter in inducing reciprocal conformity by ego. Here, as in many 
other instances of Parsonian formulations, a great deal of rigorous 
research will be required to establish the degree to which the 
abstract model— established as an ideal type or limiting case- 
usefully approximates empirical conditions. % 

With the discussion of deviance and social control we have at 
hand most of the "elements" of the model of a functioning social 

* Alvin W. Gouldner, "The Norm of Reciprocity: A Preliminary 
Statement," American Sociological Review, 25, No. 2 (1960), pp. 
172-173. For complementarity to be maintained under these condi- 
tions, Gouldner argues, a basic norm of reciprocity is necessary, over 
and above mutual reinforcement to conformity through mutual meet- 
ings of expectations and exchange of expedient rewards, 
f For a careful exploration of this feature of interaction, see Harold 
Kelley and John W. Thibault, The Social Psychology of Groups ( New 
York: John Wiley & Sons, Inc., 1959). 

I Cf. a related example of empirical rechecking in Eugene Litwak, 
"Occupational Mobility and Extended Family Cohesion," American 
Sociological Review, 25, No. 1 (1960), pp. 9-21. 


system. Yet, even in The Social System alone, one finds in addition: 
(1) an elaborate analysis of belief systems (distinguishing several 
important types, ranging from empirical science to religion); (2) 
a lengthy discussion of expressive symbols; (3) a case study of 
medical practice; (4) an analysis of social change; (5) an excursus 
on the nature and interrelations of the sciences dealing with human 
action. It is impossible here to summarize all of this material, let 
alone to go into the extensive revisions, extensions, and elaborations 
of the basic scheme which have appeared in later publications. 
However, because of the tendency of commentators to point 
especially to the structural, static, and equilibrium-maintenance 
emphases of the theoretical scheme, it seems desirable to touch 
at least upon the discussion of social change. 

The exposition departs from a distinction between processes 
within a social system and processes of change of the system itself. 
The treatment of the first set of processes has been based on the 
theoretical assumption {not an empirical generalization) that there 
is a non-problematic equilibrium of a boundary-maintaining system, 
in which interaction is assumed to tend toward stabilization of 
mutual orientations. Under these assumptions, the problem for 
analysis is that of showing how equilibrium is maintained. Hence, 
the emphasis is upon processes of socialization and mechanisms 
of social control. 

In the absence of full knowledge of the laws of process within 
the system, structural-functional theory "impounds" certain con- 
stancies of pattern into structural categories and then asks how 
these constancies are maintained or altered, and how functional 
"imperatives" limit the range of variation. Motivational processes 
are "put together" with structural factors to give descriptions of 
processes of change within the system, e.g., knowledge of family 
structure is combined with knowledge of identification processes 
to predict the development of "deviant" behavior. 

It is maintained, however, that a general theory of processes of 
change of social systems is not possible in the present state of 
knowledge; instead we can aim for theories of particular sub- 
processes. For the most part these theories, as here considered, will 
not deal with biological or physical factors, which fall outside the 
"action" schema. 

88 • Robin M. Williams, Jr. 

A general phenomenon in social change is the resistance derived 
from vested interests, arising from the integration of need-disposi- 
tions with cultm'e patterns. Such interests concern ". . . maintaining 
the gratifications involved in an established system of role-expecta- 
tions. . . ." (SS 492) Except for institutionalized change, e.g., 
scientific research, change can occur only through mechanisms 
for overcoming the resistance based on vested interests. 

On the other side, the sources of impetus of change may be a 
change in the genetic constitution of a population, in the physical 
environment, in the development of a cultural complex (science, 
religion), in technology, or in the ". . . progressive increase of 
strains in one strategic area of the social structure. . . ." There is 
no one invariant or predominant source of change. 

To analyze change, one should: (1) identify the sources of 
change; (2) identify the vested interests likely to be aflFected, or 
more broadly, describe the initial state of the system; (3) specify 
what has changed into what and through what intermediate stages; 
(4) analyze the impact of change on "functional imperatives," such 
as motivation, control of power, moral integration of the reward 
system, or cultural pattern-consistency. It is highly important to 
trace the multiple consequences of change through the system, 
paying attention to complicated "feedback" eflfects upon the original 
process of change. Many fairly detailed examples of this procedure 
are found in Parsons' writings, e.g., the analysis of the impacts of 
technological change, or of charismatic revolutionary movements. 
The merit claimed for this. approach is that changes can be located 
in relation to the detailed morphology of a social system, and the 
repercussions meticulously traced through the structure and back 
to the original point of impact. We may add that such a systematic 
inventory also opens the possibility of detecting indirect, reciprocal, 
and "mediated" ejBFects among various parts of the system. 

At the most general level. Parsons attempts to show that the 
direction of change in social systems cannot depend on the grati- 
fication-balance of individual actors. Frustration and deprivation 
because of a discrepancy between ideal pattern and actuality can 
be important in a shift from one system to another, but, it is said, 
cannot account for a continued development in a given direction. 
This point is asserted without demonstration. Evidently it depends 



upon the assumption that there is not a type of social system most 
nearly suited to universal human needs, toward which movement 
might occur by successive approximations to an "ideal fit." For 
Parsons, there is a diversity of systems in which no one type is 
clearly superior from the standpoint of optimal gratification of 
individuals. "Directionality" is to be sought, rather, in the cultural 
system. Following Max Weber, it is assumed that there is an 
inherent process of rationalization in belief systems which proceeds 
in the direction of greater rationality, unless impeded or reversed 
by forces arising from the exigencies of adaptation of personalities 
and social systems. This "tendency" is not an empiricallv observ- 
able trend; it may be compared to a potential but obstructed in- 
crease of entrophy. The directionality posited in belief systems is 
not assumed to hold for expressive symbolism, which is not cumula- 
tive but is unique to particular historical configurations. 

In the discussion of social change, then, we find the now familiar 
image of the patterning of the behavior of striving, motivated per- 
sonalities within an interactive web that is in part channelized and 
defined by cultural values, beliefs, and expressive symbols, and is 
anchored in the imperatives of biological nature and physical 
environment. The total "system" is never at rest; it is always being 
pushed, prodded and keel-hauled; it is always subject to strains 
and conflicts. Behind this relatively concrete image stands the 
abstract model of a boundary-maintaining system, "tending" toward 
equilibrium, at all levels from the exact mutuality of role-expecta- 
tions between individual actors, to the ego-integration of the 
personality to the pattern-consistency of culture. 

In this connection, a basic question concerning social systems 
and sub-systems concerns the extent to which, for purposes of 
analysis and prediction, they can be considered "closed." Our solar 
system is so far removed from other bodies that in Newtonian 
mechanics the movements of the planets can be calculated without 
significant reference to masses outside the system, i.e., it is a 
"closed system." Many physical systems are closed. In an open 
system, on the other hand, the values of the variables defining 
internal processes are more or less strongly affected by extra-system 
variables. It is undoubtedly reasonable to regard a Hving organism 
as a real system. But living organisms are to an important extent 


90 • Robin M. Williams, Jr. 

open to extra-system influences— indeed, the organism is involved 
in continuous processes of exchange with its environment so long 
as it is "alive" at all. The organism can be made a closed system 
for purposes of scientific theory just to the extent that it is possible 
to include, in predictions or postdictions about a state of the system, 
complete calculations of the influence of extra-system conditions, 
i.e., "boundary conditions." Social systems are, for the most part, 
highly open systems. Much of the difficulty of making specific 
social predictions arises from the numerous unmeasured "external" 
forces which continually impinge upon the particular system we 
have in view. Furthermore, ". . . it is easy to exaggerate the con- 
crete orderliness of modern complex societies, in all their decisive 
political and military turmoils, and this tendency is further encour- 
aged just to the extent that research focuses on enduring groups 
and upon massive formal structures. The implied challenge here 
is only to incorporate more fully and clearly in our theory and our 
research the study of such matters as discontinuities in communica- 
tion, of fluid and rapidly changing situations, of pro-normless col- 
lective behavior, of misunderstandings and lack of symmetry in 
social roles. Our world is full of crisis-conditioned, imperfectly 
structured relationships among persons and collectivities, under 
such conditions of rapid and massive change that we may require 
ideas more novel than 'equilibrium' to understand them." * 

Although change and tension are integral and important emphases 
in Parsons' thinking, it remains true that the conceptual scheme 
centers in the concept of equilibrium, and that the primary focus 
of attention is upon problems of integration. Very different emphases 
are possible, and may lead to different and important empirical 
conclusions. To take only one recent example, the work of Ralf 
Dahrendorf clearly points up the difference between an equilibrium 
model of a functionally integrated social system, on the one hand, 
and a model in which the social system is analyzed in terms of 
coercion and conflict, on the other. f Although both approaches deal 

* Robin M. Williams, Jr., "Continuity and Change in Sociological 
Study," American Sociological Review, 23, No. 6 ( 1958 ) , pp. 629-30, 
f Class and Class Conflict in Industrial Society (Stanford, California: 
Stanford University Press, 1959). 


with important aspects of actual societies, and are certainly not 
mutually exclusive, each is- subject to severe hmitations and can 
easily lead to systema.i*-- distortion. 


Efforts to utilize portions of the Parsonian schema in ob- 
servational and experimental studies have begun to test its opera- 
tional usefulness in research. In addition to the work of R. Freed 
Bales in collaboration with Parsons, an increasing number of widely 
scattered studies have made direct use of Parsonian concepts. In 
the Kansas City Study of Adult Life, a very interesting attempt has 
been made to describe the interaction style of middle-aged, aging, 
and aged persons in terms of two of the pattern-variables : specificity- 
diffuseness and affectivity -neutrality. * The dominant style of inter- 
action of "approval seekers" is specific and neutral; of "total accept- 
ance seekers," diffuse and affective. The investigators then secured 
indices of actual orientations as over against preferred orientations. 
Operational measures were devised for classifying persons in these 
ways, and satisfactory reliability of the indices was demonstrated. 
The investigators were able to show that: 

1. Men are far more likely than women to have a specific and 
neutral interaction pattern; men are somewhat more likely than 
women to prefer diffuse neutrality, whereas women are more 
likely than men to prefer diffuse affectivity. 

2. "Goodness of fit" between actual and preferred orientations 
correlates significantly with independent measures of "morale." 

3. Preference for diffuse affectivity drops sharply after age 70, 
whereas preference for both specific-affectivity and diffuse-neu- 
trality rises markedly. This change appears to involve a sloughing 
off of diffuse famihal obhgations and a greater emphasis upon 
both specific gratifications and generaHzed moral esteem. 

* Elaine Gumming, Lois R. Dean, David S. Newell, "What is 'Mo- 
rale': A Case Study of a Validity Problem," unpublished paper 

92 • Robin M. Williams, Jr. 

The Parsonian formulations have influenced such studies of 
role conflict as those of Toby, Stouffer, and Getzels and Guba.* 

The substantial monograph of Gross, Mason, and McEachern is 
permeated by discerning and critical utilization of ideas drawn 
from The Social System.-\ 

In a study of friendship choices in relation to similarity and 
dissimilarity of values, the pattern-variables were used to index 
"friendship." J Although this was an exploratory study, used as 
a research-training exercise, and although the measures were 
crude, the results of the attempt to give operational specifications 
of friendship interms of the pattern-variables represent an advance 
over methods previously used in research on this subject. 

These few examples are representative of a larger number of 
studies in which Parsonian concepts are being employed. In addi- 
tion, of course, the scheme has exerted substantial influence upon 
works of analysis and interpretation such as those by B. Barber, 
E. Devereux, K. Davis, W. J. Goode, H. M. Johnson, M. Levy, 
R. K. Merton, W. E. Moore, R. Wilhams, L. Wilson, and many 


In the beginning let us dispense with those criticisms 
which concern the style of presentation of the conceptual scheme. 
We readily grant that neologisms abound, that sentences sometimes 
appear to be literal translations of a text originally written in 

* Jackson Toby, "Some Variables in Role Conflict Analysis," Social 
Forces, 30 (1952), pp. 323-327. Samuel A. Stouffer, "An Analysis of 
Conflicting Social Norms," American Sociological Review, 14 ( 1949 ) , 
pp. 707-717; Samuel A. Stouffer and Jackson Toby, "Role Conflict 
and Personality," American Journal of Sociology, 55 (1951), pp. 395- 
406. J. W. Getzels and E. G. Cuba, "Role, Role Conflict and Effec- 
tiveness," Amencan Sociological Review, 19 (1954), pp. 164-175. 
f Neal Gross, Ward S. Mason, and Alexander W. McEachern, Ex- 
plorations in Role Analysis (New York: John Wiley & Sons, Inc., 

I Robin M. Williams, Jr., "Friendship and Social Values in a Sub- 
urban Community: An Exploratory Study," The Pacific Sociological 
Review, 2, No. 1 (1959), pp. 3-10. 


German, that the style is complex, that the use of terms is not 
always consistent, and that some passages still defy comprehension 
after repeated and earnest scrutiny. Nevertheless, there is an intel- 
lectual content which can be grasped, and it is that content in 
which we are interested. 

The Parsonian "system" is not a unified deductive system. When 
it is said that a certain concept is "derived from" a previously 
defined concept, it is only rarely that a strict logical derivation is 
found. More usually the new concept is (1) a connotative deriva- 
tive, which is developed by explicitly defining a connotation of the 
original notion, or (2) a linking term which is introduced between 
two concepts, or (3) a concept derived by a cross-tabulation of 
independently defined constructs, or (4) a term developed to talk 
about phenomena empirically associated with the referents of the 
original concept. Although one can find deductive chains of reason- 
ing at various points, no major portion of the work is a postula- 
tional system such as characterizes deductive economics. 

At the most general level the Parsonian treatment forthrightly 
summarizes several highly important, if very general, assumptions 
concerning man and society that are supported by a large amount 
of evidence. For example: 

1. A large amount of human social action is goal-directed. 

2. Social action is sufficiently patterned to allow for analysis in 
terms of systems. 

3. As the only symbol-using animal, man is able to generalize 
from experience and to stabilize a pattern of behavior through 
time. Simple stimulus-response interpretations are inadequate to 
account for these facts.* 

4. Action is, in part, directed by orientation to value-standards. 

5. Action-systems represent "compromises" among organismic, 
cultural, personality, and social systems, as motivated actors con- 
tend with the exigencies of survival in an environment. "Perfect 
integration" is not found in the empirical world. 

Even though uneven in logical development, one of the virtues 

* ". . . the high elaboration of human action systems is not possible 
without relatively stable symbohc systems where meaning is not pre- 
dominantly contingent on highly particularized situations." (SS 11) 

94 • Robin M. Williams, Jr. 

of the Parsonian conceptual system is its comprehensiveness. As 
Devereux has pointed out, it provides an elaborate checklist, as 
it were, for the description of any social system; as one uses the 
system he is continually reminded to look for structures and proc- 
esses which might go unremarked were not some systematic guide 
being followed; he is encouraged to trace in detail the structural 
elements of the society, the modes by which conformity is elicited, 
the sources of deviance, the formation of adaptive structures, the 
sources of change and of resistance to change. It is in this way 
that the scheme not only facilitates description but also can serve 
as a diagnostic tool. For Parsons keeps asking at every point, 
"What can 'go wrong' here?" and, "How does the system cope with 
'trouble'?" His interest is not exclusively in a static inventory of 
characteristics of culture, personality, and society, but in an analysis 
of dynamic functioning of interlocked systems which are always 
imperfectly integrated, always subject to strain. "Equilibrium," or 
order, is always empirically problematic; it is not a global emana- 
tion but the specific outcome of highly complex interrelations among 
specific processes. 

We are here prepared to reject two sharply opposing views of 
Parsons' use of the concept of "equilibrium." In the full context of 
its usage in his several major works, "equilibrium" does not mean 
to Parsons that real societies and real systems always tend to 
"correct" deviations immediately, and with minimal and slight effort, 
and return to a steady state. This notion is repeatedly and emphat- 
ically rejected, as is the notion that empirical social systems can 
ever be perfectly "integrated" at any one time. Furthermore, 
throughout the later works there is recurrent and specific concern 
with strain, deviation, alienation, and "compulsive" conformity. On 
the other hand, the critics seem to have made valid points, which 
cannot be summarily dismissed as results of careless reading or 
inadequate comprehension. Even with all its careful disclaimers 
and quahfications, the scheme does have the net effect, for many 
readers, of emphasizing stability, and, by omission, understating 
the problem of radical discontinuities and rapid, massive, and vio- 
lent conflicts and changes in social systems and sub-systems. Thus, 
the idea of equilibrium has both intriguing possibilities of system- 
analysis and disconcerting difficulties of operational definition and 


of eflFective employment in empirical research. We would not sup- 
pose it useful to take as an equilibrium-state any momentary given 
condition of a group or society— whether the "moment" in question 
be one-half hour in a small-group laboratory experiment or a 
decade in the history of a nation. For the concept of equilibrium 
to have descriptive, predictive, or explanatory value, it must be 
possible to state a set of defining conditions specifying what 
*l3alance of forces" is to be "equilibrium" and what "movements"— 
change in previously selected relevant variables— will constitute 
disequilibrium. This, we submit, Parsons has not done. The con- 
cept of equilibrium, like the related term, integration, still floats 
freely in the high reaches of "free intellectual creation." A particu- 
larly agonizing question here, as in most other major portions of 
the conceptual scheme, is whether the Parsonian scheme can 
generate real predictions or is restricted to post facto classifications 
and interpretations.* 

The conceptual framework we have reviewed probably comes 
closer than any other modern synthesis to an actual conceptual 
linkage of considerable parts of anthropology, economics, political 
science, and sociology. Although the linkage is highly abstract, it 
does serve to place the several social sciences in a new perspective, 
and at least points in a very general way to more specific areas of 
theoretical and empirical articulation among these disciplines. 

Parsons explicitly disavows any intention to present his work as 
a theory of concrete social phenomena. f Contained in his writings 
are several quite different kinds of attempted contributions : ( 1 ) 
critiques of other conceptual schemes and theories; (2) general 
methodological analysis; (3) development of new constructs; (4) 
social taxonomy (structural classification); (5) translation of one 
conceptual scheme into another and specification of relationships 
between nontranslatable concepts (theory of social action, and 

* And further, if predictions can be derived, whether the data re- 
quired are too massive and fugitive to justify the effort required, i.e., 
would we have to know "too much" in advance in order to make 

f "It is not an attempt to formulate a theory of any concrete phe- 
nomenon, but it is the attempt to present a logically articulated con- 
ceptual scheme." (SS 204) 


96 • Robin M. Williams, Jr. 

Freudian concepts); (6) numerous observations and fresh insights 
concerning actual social behavior; (7) broad diagnoses of social 
situations (Nazism in Germany, An;ierican kinship systems); (8) 
hypotheses suggestive of research needs; (9) certain empirical 
generahzations. This almost certainly is not a complete list, but 
it is enough to warn us that a critique of the whole edifice must 
be careful to specify its points of reference. 

In a complex and rapidly developing field of study, debates con- 
cerning strategies for advancing knowledge are inevitable, and 
questions about the long-run fruitfulness of alternative modes of 
attack are not likely to be put to rest until the actual returns are 
in, if then. Yet it is neither unnecessary nor unprofitable to have 
such debates go on. The confrontation and comparison of "con- 
ceivable futures" for research and theoretical construction often 
clarify the possible choices, narrow the area of disagreements, 
and in a variety of other ways help to refine and temper our judg- 
ments. Parsons elaborates an impressive case for the advantages 
of comprehensive and abstract conceptual schemes. Merton advo- 
cates middle-range formulations, dealing with coherent subfields 
of sociological problems. And a large number of research workers 
are busily engaged in describing social phenomena or testing 
specific hypotheses with only remote and tenuous connection with 
any explicit "general theory." Although hardly anyone engaged in 
social science research these days would seriously argue for old- 
fashioned "raw empiricism," the spectrum of opinions as to effective 
strategy is obviously wide. 

Conceptual schemes may be developed, as Sheldon puts it, as 
free creations of the human intellect. Or, they may be devised in 
relatively close relation to observation and experiment, growing 
by continuous revision and extension as new data force changes 
and as the changing concepts lead to new data. For the most 
part Parsons has elected to drive directly for a comprehensive and 
systematic conceptual scheme; presumably this scheme is supposed 
to guide subsequent research in producing the data needed to give 
specific empirical content to the categories and hypotheses. Cer- 
tainly this approach is understandable as a reaction against raw 
empiricism and ad hoc conceptualization. On the other hand, it is 
clearly true that Parsons gives very few hints as to how the con- 


cepts could be given an operational meaning. If the scheme is to 
be of any use in research, the main operational task lies ahead. In 
this reviewer's judgment, serious efforts to use components of the 
scheme in research are likely to lead to quite substantial modifica- 
tions of the conceptual apparatus itself. 

Such modifications will probably be encountered, for instance, 
in first-hand research on the relation between "shared value-orienta- 
tions" and "mutually adjusted role-behavior." First of all, can the 
two sets or complexes of variables be defined in a nontautological 
manner, that is, by independent operations? If they do prove to be 
amenable to such definition, a host of further questions arise, viz.: 
What value-content is to be indexed, and how? What specific 
aspects of mutual adjustment should be included? What particular 
hypotheses are to be investigated? Shall one attempt to secure 
measures of intensity and salience of value-commitment? Do we 
rely upon testimony, projective tests, or direct observation of 
behavior? And so on. 

Or, suppose we wish to determine the part played in the "soli- 
darity" of a religious collectivity by "common values." What values? 
How indexed? Assuming, as is likely, that we find or devise several 
different operational indices of the variables, how do we select 
from them or combine them? Is "solidarity" merely another way 
of saying "shared values," or can we find reasonably specific 
independent measurements? 

Again, what does it mean to speak of status as a "position"? It 
seems that as a concept referring to social reality a "position" can 
only mean a certain organized cluster of rights and obligations 
attributed by alters to an ego. A status may exist without being 
named or explicitly recognized. The fact that a status is named, 
e.g., "father," really tells us nothing directly of the rights and 
obligations which define what a "father" is expected to do. Parsons 
thinks of status as "... a place in the relationship system considered 
as a structure, that is a patterned system of parts." (SS 25) But: 
"It is difficult to separate the idea of location from the relationships 
which define it . . . persons cannot be located without describing 
their relations to other individuals; the points imply the relation- 
ships and the relationships imply the points. . . . Since positions 
have been defined as locations of actors in systems of social rela- 

98 • Robin M. Williams, Jr. 

tionships, they can be completely described only by an examination 
of the content of their interrelationships." * Here, once more, 
operational criteria are crucial. It cannot be taken for granted that j 
even the most commonly recognized, and seemingly obvious and 
definite, statuses, do in fact "exist" in terms of clear definition by 
overwhelming consensus of a population. 

Gross et al., in the work just cited, struggle diligently with the 
problem of actually indexing the "roles" of school superintendents 
in the State of Massachusetts. Before they feel prepared to study 
this single role they find it expedient to develop the following 
terms: position, positional sector, expectation, role, role sector, 
right, obligation, role behavior, role attribute, role behavior sector, \ 
role attribute sector, and sanction.^ It is an instructive exercise 
to follow their efforts to secure actual indicators for these concepts 
in one relatively restricted empirical study. 

One other illustration will sujffice to document the point now 
under discussion. In The Social System, Working Papers, and 
Family, Socialization and Interaction Process, there are frequent 
references to processes which go on in psychotherapy; these are 
summarized as permissiveness, support, denial of reciprocity, and 
manipulation of rewards. These four processes are regarded as 
suggestive of important aspects of socialization and, indeed, of 
social interaction generally. Now the discussion of these processes 
is illuminating, perceptive, and stimulating, and possessed of con- 
siderable persuasive appeal. Yet the concepts as set forth do not 
readily or obviously anchor themselves to observable behaviors. 
As a matter of fact, a great deal of research has already been 
directed to the analysis of the psychiatric interview; it is not sur- 
prising that these efforts show that indices of process that are 
objective, that can be subjected to intersubjective confirmation, are 
difficult to devise, but it is important to note that no one con- 
ceptualization of these processes appears as yet to have demon- 
strated superiority as a basis for research. 

In my view, a theorist is entitled to say: here are important 
ideas; it is not my task to say just how you use them in research 

* Gross, Mason, and McEachem, Explorations in Role Analysis, p. 48. 
f Ibid., p. 67. 


operations. On his side, the empirical research worker would seem 
to have the right to suspend judgment on the scientific usefulness 
of any conceptual scheme until serious efiFort to employ it has 
shown high productivity or low productivity of empirically tested 
hypotheses. My conclusion is that the full returns are yet to be seen 
and that a definitive judgment on the overall merits of Parsonian 
sociology cannot now be made. Even while saying this, I recognize 
that two further fairly significant appraisals are possible even at 
this time: (1) that the system has demonstrated a high degree 
of provocative value in stimulating other workers to examine data 
in the light of this conceptual scheme, and that important results 
have thereby been obtained; (2) that the empirical usefulness 
of the scheme, thus far, has been in the application of limited 
parts of it to the interpretation of data, to the development of 
hypotheses, and to the descriptive ordering of information about 
particular institutions and societies. 

There is no doubt at all that the work of Parsons stands as a 
massive intellectual achievement— perhaps on the whole the most 
widely recognized theoretical work of any contemporary sociologist. 
In closing these critical comments, it is appropriate to note also that 
among the students of Parsons there have been very few "disciples": 
those who have used his contributions have used them, in the main, 
selectively and critically. In so doing they often have modified his 
views, and by challenging his concepts and assumptions have not 
infrequently been stimulated to productive research and scholar- 
ship. It is a plausible guess that these reactions partly reflect the 
rigorous critical standards and continuous lively curiosity discern- 
ible in the works here reviewed. 


I THE , 

Chandler Morse 



No one familiar with economics can fail to remark the 
resemblance between many Parsonian concepts and those 
of the older discipline. This is hardly strange. If, as Parsons 
contends, economics is only a special case of general action 
theory, but also a very important special case, the two 
systems are interdependent. 

Parsons' approach to economic questions is much broader 
than that of orthodox economics, though not broader— 
and in some respects narrower— than dissenters have con- 
sidered desirable. The names of Marx, Hobson, Roscher, 
and Veblen stand out among a host of critics belonging to 
the socialist, German historical, and American institutional 
schools who have attempted to widen the boundaries of 
the discipline. But apparently it won't do to try to treat 
noneconomic matters as mere modifiers of a theory that is 
committed by its essential nature to expounding the grounds 
of economic behavior and exploring its consequences. 

* This paper is an outgrowth of an exploratory study of institutional 
change and economic growth undertaken in collaboration with Charles 
Wolf, Jr., with the aid of a grant from the Ford Foundation. Neither 
shares responsibihty for its content. 


Parsons, by undertaking to treat economics and the other speciahzed 
social discipHnes as aspects of a general theory of social action, has 
removed one obstacle to progressing toward the goal visualized 
by the dissenters— but he has acquired a host of problems in 


What the Model Is About 

The Parsonian approach rests on the familiar premise that the 
complex a£Fairs of a society could not be conducted unless they 
were organized in some systematic way; and on the further hypoth- 
esis that human societies, from the most primitive to the most 
complex, have so much in common that there must be a set 
of fundamental organizing principles shared by all societies but 
carried to much higher degrees of elaboration in some than in 
others. The aim is to discover what these basic principles are 
and how they operate. 

The cross-cultural similarities of social organization and process 
arise because, given the nature of the human organism and the 
physical environment, certain "problems" must be solved if man 
is to live as a social animal— that is, to employ scarce means 
cooperatively (socially), and more or less rationally (sometimes, 
economically) to attain given ends. What these problems are, 
says Parsons in eflFect, can be determined by analyzing the require- 
ments of this cooperative and (more or less) rational ends-means 

The first step is to recognize that the mere statement of what 
the process involves identifies two analytically distinct problem 

(1) That of executing technically effective methods of coopera- 
tion for attaining ends approved by the cooperators. 

(2) That of maintaining efficient cooperation. 

Looked at from one point of view, these problems seem to be 
mutually independent, the first calHng for solution of essentially 
intellectual problems, the second for solution of essentially emo- 

102 • Chandler Morse 

tional problems. From another point of view, they seem not to 
be distinct problems at all; for if the ends of cooperation are 
approved by the cooperators, why shouldn't cooperation continue 
indefinitely? (That it should, incidentally, is implicit in the neo- 
classical economic model.) 

Parsonian theory implies that cooperation always tends to 
break down. Since few of the ends of activity are "final," most 
being simply means to more remote ends, there is a perpetual 
problem of preventing what Parsons calls "premature gratifica- 
tion" (WP 184), of keeping noses to the grindstone. This causes 
"strain" and is one reason why systems don't maintain themselves 
automatically. In addition, the roundaboutness of the means-ends 
circuit inevitably results in differential consequences for the co- 
operators (e.g., differential sharing of the costs and benefits of 
cooperation, which raises the question of social justice). Unless 
integrative measures can render differential consequences accept- 
able, or the social system can provide coercive structures to deal 
with nonacceptance, cooperation will not persist. (WP 211) 

Consequently, there really are two problems, as stated initially. 

Much trouble would be saved if this were all there were to 
the matter. Difficulty arises because the two problems are neither 
neatly distinct nor identically one. They are intermeshed in the 
tissues of the human organism, where intellect and emotion lead 
a blended but conflictful existence.* If this explains the complex- 
ity of Parsonian theory we had best make up our minds to put 
up with it. 

Unfortunately, the inherent problem posed by the blending of 
our objective and our subjective lives— to use the two old-fashioned 
terms— is not the only reason for complexity. A further reason is 
that Parsons has either not always been clear concerning the 
nature of the difficulty, or (more probably) has kept changing his 
mind concerning the way out of it. The result is that the concepts 
of the General Theory and The Social System do not wholly fit 
with those in Working Papers, and undergo further changes in 

* On this, see V. J. McGill, Ernotion and Reason (Springfield, 111.: 
C. T. Thomas, 1954). 


Family and in Economy and Society. Clarification of the gratuitous 
obscurities is badly needed. 

At the outset, it appears, Parsons undertook to think of the action 
process as involving two essentially distinct types of activity, 
together with a third type that somehow spread-eagled the other 
two. The first of the basic types of activity he called "instrumental," 
by which he apparently meant activity devoted to solving the 
cognitive problems of the means-ends network; the second type 
he called "expressive," apparently meaning activity designed to 
solve the cathectic problems.* 

This would have done very well if it had been possible to stop 
there. One could simply have held that the problem of finding 
means called for instrumental activity, that of choosing ends for 
expressive activity. But there was still the problem of maintaining 
cooperation, which Parsons viewed initially as calling for integra- 
tive activity, a third basic type. No reader of Chapter III in The 
Social System can fail to be aware that Parsons was having con- 
siderable diflBculty fitting this third type into his model. The 
reason was that the basis of classification was different. Whereas 
the first two types of activity were classified according to their 
source within the personality, the third was classified in terms of 
what it was meant to accomplish. Consequently, integration became 
a special kind of end that called for the employment of special 
kinds of means. It therefore was a special case of the ends-means 
relationship, calling (perhaps) for both instrumental and expres- 
sive activity. 

Out of this confusion there emerged the model that took shape 
in Working Papers and was applied and extended in Family and 
Economy and Society. The model's general design, and especially 
its development, appear to have been influenced heavily by efforts 
to deal with four basic questions. Parsons does not explicitly 
formulate these questions but they seem to be implicit in his ap- 
proach. The questions are: 

1. What are the instrumental (intellectual, cognitive, objective) 
interests, and what are the expressive (emotional, cathectic, sub- 

* "Cathexis, the attachment to objects which are gratifying and re- 
jection of those which are noxious, lies at the root of the selective 
nature of action." (GTA 5) 

104 • Chandler Morse 

jective) interests, that are represented in any behavioral process, 
and how do they operate within or upon the process? 

This question underUes the distinction made in Working Papers 
between "object-system process" and "motivational process." 

2. What type of value does a given process create— that is, does 
it create a "product" that is valued as a means or one that is 
valued as an end (or one that is valued both as means and end, 
such as personal wealth). 

This is the basis for the distinction between "facilities," which 
are valued as means, and "rewards," which are valued as ends. 
Both facilities and rewards can be physical objects, cultural (in- 
cluding symbolic) objects, and (rarely) social objects, meaning 
individuals or collectivities. An object can be both facility and 
reward: Viewed as a facility its significance is instrumental, as 
a reward, expressive. 

3. What is the ultimate significance of a given process from the 
system point of view? Taking the system as a normally functioning 
organism, can we say that the process is one that focuses on 
"getting the job done" and promotes the attainment of a valued 
goal state? Or, alternatively, taking the goal states of the system 
as provided for, can we say that the process is one that will help 
to maintain the system as a normally functioning organism? 

This is the basis of the distinction between "task-performance" 
and "system-maintenance." 

4. Is a given process to be viewed from the standpoint of the 
system as a whole or from that of a given unit or sub-system within 
the system? 

This is the problem of "system reference," which crops up con- 
tinually. It is especially important as regards (a) Goals: are they 
"ultimate" or "intermediate"? (b) Processes: are they inter-system 
or intra-system, that is, inter-unit? (c) Socio-emotional needs: are 
they those of units of a system taken by themselves or of a system 
as a whole— of units in relation to other units? In the latter con- 
nection we must always remember that when we look inside any 
"unit" it is seen to be a system— though diflFerent in kind from 
that to which the unit belongs; and that every system can be 
regarded as a unit (sub-system) of a more inclusive— and diflFerent 


The two following sections undertake to ferret out the hard core; 
of Parsons' exphcit or implicit answers to these and certain related 

System Structure and Action Process 

Some Basic Concepts. A process beginning with goal-directed 
behavior and ending with attainment of the goal is an action cycle; 
it occurs within a "system of socal action." A system of social action 
is composed of three types of sub-systems: (1) the personality 
systems of at least two individual actors, consisting of internalized 
"need dispositions" and therefore of potential "motivational com- 
mitments" to various types of goals and to various patterns of 
behavior; (2) the social system, or structure of social organization, 
consisting of defined roles and their associated and institutionalized 
( = internalized and shared ) role-expectations ( = expected "per- 
formances" and "sanctions"); and (3) the culture system, consisting 
of the heritage of knowledge, beliefs, ideas, technologies, mores, 
customs, habits, laws, values, standards, norms, together with the 
symbols, both tangible (artifacts) and intangible (language, the 
arts) that represent them. 

No one of these systems is entirely independent of the others. 
The culture system is the major binding element. Certain cultural 
elements are embodied in personality systems, others in social 
systems, and still others are "out there" as a sort of perpetual 
stock, analogous to natural resources, that is always available for 
use in action processes. No system of action can survive unless 
these three aspects of culture are mutually consistent within some 
degree of tolerance. Need dispositions can be thought of as the 
organization chart of individual action systems; role-structure as 
the organization chart of social action systems; and culture as the 
organization chart of organization charts. (GTA 4-27) This is the 
basic structure of social action, and makes a basic (though insuffi- 
cient) contribution to the stabilization and coordination of action 

An action system also includes "objects," that is, concrete- 
though not necessarily tangible— things such as actors, physical 
objects, cultural objects (including symbols), personality systems, 
social systems, and other systems of action (including the reference 

106 • Chandler Morse 

system itself as object). This is the "situation of action," or object 
world, which has certain properties that will be perceived as 
significant for action purposes. In general, these properties are of 
two kinds: cognitive, what the object is, and cathetic, what the 
object is good for (or bad for). It is a major attribute of the cul- 
ture system that it provides a certain measure of assurance that 
all individual members of a system of action will, except within 
the acceptable limits of tolerance of idiosyncrasy, "see" objects in 
essentially the same terms, as possessing essentially the same signif- 
icant properties. Indeed, a posited condition of the survival of 
systems of action is that the members be bound together and their 
behavior patterns be stabilized and coordinated in this way to a 
rather high but unspecified degree. 

It is not sufficient for members of action systems to share cogni- 
tive and cathectic standards to a degree; they must also share 
evaluative standards— the standards by which, when confronted 
with the necessity of making a choice among alternatives of any 
kind, the alterantives can be rank-ordered according to each rele- 
vant dimension and then, possibly, according to some measure 
of their aggregate value.* The effective pursuit of system goals 
requires that such evaluations by members be mutually consistent 
(within a tolerable standard of variance). This is another basic 
source of stabilization and coordination. 

Roles and Transactions. Roles are job descriptions for the slots 
specified by the organization chart of society. They define the 
institutionalized obligations and rights of the role occupants— their 
areas of responsibility and authority. To what extent, and with what 
degree of precision, the goal-orientations, or functions, of roles 
must be specified Parsons does not say; but that they must be 
specified seems to be implicit in the entire analysis. 

The responsibilities of roles fall into two broad categories. f The 

* Parsons maintains such strict ethical neutrality that he fails to in- 
quire into the determination of evaluative standards. He does not 
explore the psychological foundations of ethics, or of the concept of 
social justice, and therefore has nothing to say concerning possible 
basic guidelines of the process of social evolution, 
f Parsons' use of the term "role expectations" places emphasis on 
rights. The term "responsibilities," which emphasizes obligations, has 
a more positive character. 


first specifies how the role occupant must manipulate objects— that 
is, operate upon them so as to change their properties, their loca- 
tion, their relation to other objects, and so on. This is technical 
role behavior. 

Parsons has little to say about the technical aspects of roles. 
In specifying their existence, and in concerning himself with the 
problem of operating upon the environment in the interest of goal 
attainment, he recognizes the central importance of manipulating 
objects. Object manipulation, being most of what people really 
"do," is what we naturally think of as the central feature of action 
process. But such manipulations are of subordinate concern to 
Parsons, for they are incidental to the processes of interaction, 
which is what the theory is about. That is why Whyte can remark 
that there is "no action" in Parsons.* The productive processes of 
adapting means to ends are fundamental to Parsons' model, but 
the model itself is more concerned with interactions and trans- 
actions than with manipulations and productive transformations. 
The latter go on, however, and if this be remembered much that 
seems rarefied and abstract becomes more solid and concrete. 

The second category of responsibilities specifies how the role 
occupant shall interact with others— what performances and sanc- 
tions he shall render. 

In the process of interaction, an act analyzed in terms of its 
direct meaning for the functioning of the system, as a "contribu- 
tion" to its maintenance or task performance, is called a per- 
formance. On the other hand, an act analyzed in terms of its 
efi^ect on the state of the actor toward whom it is oriented (and 
thus only indirectly, through his probable future action, on the 
state of the system) is called a sanction. This is an analytical 
distinction. Every concrete act has both a performance and a 
sanction aspect. (ES 9) 

The economic distinction between supply and demand is cited 
as a case in point. Supply focuses on providing the economy as a 
whole with particular classes of goods and services; it is thus 
a social performance. Demand focuses particular requirements for 
various goods and services on the particular supplying agencies; 

* See p. 255. 

108 • Chandler Morse 

it thus influences their wiUingness to maintain or increase supply, 
and is a social sanction. (ES 10) 

The simplest but by no means the only case of interaction 

... is that of reciprocity of goal orientation, the classical eco- 
nomic case of exchange, where alter's action is a means to the 
attainment of ego's goal, and vice versa. (SS 70) 

In highly organized and durable systems of exchange egos tend 
to become specialized in producing means for the attainment of 
the goals of others. Reciprocally, the attainment of ego's goals 
becomes enmeshed in expectations of what are (to them) signifi- 
cant actions by alters. 

What an ego gets thus depends not only on what he produces 
but also on the "terms of exchange" of his products for alter's. 
(GTA 210 ff. ) It is not necessary, however, that every ego-alter 
relationship be of the kind properly called an exchange in the 
technical economic sense of involving a quid pro quo. Ego may 
transfer product to an alter from whom he expects to receive 
nothing of specific equal value in exchange; and he may receive 
transfers from other alters on the same basis. A transfer involves 
no explicit quid pro quo. 

Following usual economic practice, exchanges and transfers 
will be referred to here as transactions. Transactions are thus proc- 
esses that consist of flows between units or systems, together with 
the activities directly involved in settling the terms governing the 
flows. Processes internal to a unit or system, that either precede or 
do not involve flows between units or systems, will be referred to 
as transformations. (WP 216) 

The analytical distinction between exchanges and transfers is 
not made explicitly by Parsons but his analysis implies that ex- 
changes occur when, and only to the degree that, the transacting 
parties belong to what, for the purpose of the transaction in ques- 
tion, are different solidary systems, that is, different "collectivities"; 
and that transfers occur when, and only to the extent that, the 
parties regard themselves as belonging, for the purpose of the 
transaction, to the same solidary system. Put otherwise, exchange 
implies "self-orientation" on the part of one or both parties, and 
transfer implies "collectivity -orientation" on the part (at least) of 


the transferer.* Thus, where transfers occur, diffuse claims and 
roundabout expectations of eventual sanction replace specific and 
direct sanctions. Since solidarity is a matter of degree, varying in 
both scope and intensity, these distinctions are also matters of 
degree. Concrete transactions may include both exchange and 
transfer elements in varying proportions. 

Transactions call for certain types of performance-sanction inter- 
change. In addition there must be something which changes hands, 
which is disposed of and received. 

This something may be control of a physical object in certain 
respects, including power to desti'oy it (e.g., food through "con- 
sumption"). It may be an agreement to do certain things in the 
future, positive as contributing to alter's goals, or negative as 
refraining from interfering with alter's goals. This something 
will be called a possession. (SS 71) 

The social significance of possessions arises from and is embodied 
in the fact that they are bundles of rights and obligations. 

Possessions come into being as the result of action processes; 
they are the "products" of action, and so are included among the 
"consequences" of action. 

Possessions are of two types, facilities and rewards. Facilities in 
their tangible aspect are such things as materials, equipment, real 
estate— items to be used to attain some goal and not as objects of 
direct gratification. Whether labor services are to be regarded as 
a facility is uncertain. It would seem that they should, but their 
handling in Economy and Society suggests that they are not. 

Rewards are possessions that have an expressive ( gratificatory ) 
significance to the recipient. Rewards and facilities may be the 
same concrete possessions viewed in two different ways, but they 

* "It became apparent that [self -orientation and collectivity-orienta- 
tion] . . . defined the relations between two systems placed in a 
hierarchical order. Self -orientation defined a state of relative inde- 
pendence from involvement of the lower-order in the higher-order 
system, leaving the norms and values of the latter in a regulatory, 
i.e., limit-setting relation to the relevant courses of action. Collectivity- 
orientation on the other hand defined a state of positive membership 
whereby the norms and values of the higher-order system are posi- 
tively prescriptive for the action of the lower." (ES 36) 

110 • Chandler Morse 

may also be distinct objects. In any case, "rewards are always to 
be understood as part of the complex of expressive symbolism, not 
part of the instrumental means-ends complex." (SS 119) Rewards 
may be either positive or negative (i.e., penalties). 

Stabilization of Interaction Process. Transactions are inherent in 
the very nature of social systems. The basic stabilizing and coordi- 
native devices discussed earlier cannot handle the infinite variety 
of possible transactions. Special mechanisms are therefore needed 
to assure that the terms on which transactions are settled will, 
in general, be compatible with the stability of the system. In The 
Social System, two mechanisms are provided for the orderly set- 
tling of the terms of exchange between (competing) members of 
diflFerent sub-systems: an "economy of instrumental orientations" 
and a corresponding "economy of expressive orientations." The 
former is concerned with the problem of allocating rights to facil- 
ities, the latter with rights to rewards. This distinction appears not 
to have been employed in later work. 

Transactions are the substance of interaction process; they con- 
sist of a reciprocal discharge of role responsibilities. If we consider 
any two related roles the implied interactions may be summarized 
as follows: 

1, a. Performances by ego that contribute to disposal of his "prod- 
uct" and to providing alters with facilities (which may also have a 
reward significance to alter). 

1, b. Sanctions by alters that reward ego for his performances 
(where the rewards may also be significant as facilities for ego). 

2, a. and 2, b. Performances by alter and sanctions by ego. 

The "circular flow" that economists have long recognized to be 
an inherent characteristic of economic activity, and to be respon- 
sible for the complex interdependencies of modern societies, is 
thus represented as a particular case of a circularity that is inherent 
in all social action. 

Markets, which are governed by the institution of contract 
(ES 104-113), are the standard mechanisms for regulating the 
settlement of terms in competitive, nonsolidary, "ecological" sys- 
tems. (SS 93) An ecological system is one in which the transactions 
are those of exchange. For the purpose of the exchange, the actors 
view each other as belonging to different collectivities, as bound 


only by ties of mutual interdependence. Competition, supplemented 
by a variety of permissive and prescriptive rules, is the agency 
through which rights to possessions are allocated in a market. Out- 
side the market context, and for other purposes— such as those of 
a social club or a nation— the actors who compete for some purposes 
may think of themselves as belonging to the same solidary system, 
that is, to the same collectivity. 

Within a collectivity role relationships are cooperative, not com- 
petitive. Transfer, not exchange, is thus the appropriate mode of 
allocating rights to possessions. Conformity of role behavior to 
expectations, therefore, is not assured by a bargaining process, in 
which the value of sanctions is brought into line with that of 
performances, and another mode of regulation is needed. This is 
made possible by the existence of loyal attachments among the 
members of the collectivity and especially by "solidarity," which 
comes about through the institutionalization of loyalties. 

Even in a solidary system it is impossible to expect appropriate 
role behavior at all times. Deviance in various forms is endemic 
to social systems. To maintain the system as a going concern in 
the face of distintegrative behavior requires special mechanisms 
of social control. (SS ch. VII) As these are peripheral to action 
process per se they will not be dealt with here. 

Coordination of Role Behavior. Stabilization of role behavior is 
necessary for continued system functioning, but it is not sufficient. 

A given action process or cycle, focused on attainment of a 
specific goal, is a complex of many articulated processes that must 
be adapted to a great number of unforeseen circumstances ( "exigen- 
cies"). Somehow, therefore, these processes must be coordinated. 

The need for close coordination is most clearly seen in an 
organization. An organization is a "system of cooperative relation- 
ships" (SS 72) capable of "continual action in concert" (SS 100), 
and having "primacy of orientation to the attainment of a specific 
goal." (ASQ 64) The latter characteristic requires that coordina- 
tion be achieved in large degree by explicit and formal means; the 
implicit, informal methods of unorganized cooperative systems 
would provide too little assurance that the specific goals, pursuit 
of which is the raison d'etre of organizations, would be effectively 
and eflBciently attained. 

112 • Chandler Morse 

Parsons pays little attention to the problem o£ explicit coordina- 
tion;* implicit in his approach, however, there appear to be the 
following methods of achieving it:' 

1. Competition, as in exchange transactions. 

This operates when actors regard themselves as belonging to 
different systems and as pursuing different, though presumably 
consistent or reconcilable, goals. It is the process of reconcilation 
that achieves coordination. The market is the prototype of this 
method. The final arbiter of competition is bargaining power ( which 
should be in balance for best results). (ES 146) 

2. Collaboration, as in transfers. 

Actors who, while interacting, regard themselves as members of 
the same solidary system, as pursuing common goals, adopt the 
collaborative mode of coordination, that is, the method of mutual 
voluntary adjustment. Collaboration is necessary to all organiza- 
tions. The family is a clear case of a cooperative system that is 
not an organization. The final arbiter of collaboration is authority 
( legitimized superior power ) . 

3. Coercion. 

When the attainment of any collectivity's goals requires the tech- 
nical cooperation of actors who are not regarded (or do not regard 
themselves ) as members of the collectivity— i.e., as sharing its goals 
—they may be coerced into involuntary cooperation. The essential 
requirement for coercion is an unbalance of power. 

These three analytical modes of coordination are blended in 
concrete situations. No continuing pattern of cooperation can be 
maintained in a complex system without employing each of the 
three methods of coordination in some measure. Yet social systems 
may often be distinguished according to the degrees in which 
they employ the several methods for specific purposes. The USSR 
treats production as a national goal, and its economy is an organiza- 
tion of national scope. Competition contributes scarcely at all to 
economic coordination. Production in Western societies is seldom 
organized much beyond the individual enterprise level; within the 
enterprise competition is unimportant but as between enterprises 
it is the main coordinative device. 

See Section VII, pp. 149-52. 


The three indicated modes of coordination do not, perhaps, 
exhaust the possibihties. Must competition be construed broadly to 
inckide conflict? Must voting? Do arbitration and judicial process 
belong with collaboration? Or with coercion? Or with both? What- 
ever the appropriate answers, there is presumably a relatively 
small number of coordinative modes (though some that are pos- 
sible may not yet have been invented). Is there some criterion 
for determining the optimum mode or combination of modes for 
each situation? 

Basic Design of the Model: Four Functional Problems 

The relation of structure to process was far from clear in early 
versions of the Parsonian model. But as the model evolved, the 
relationship acquired an increasingly definite form, based on the 
hypothesis that 

. . . process in any social system is subject to four independent 
functional imperatives or "problems" which must be met ade- 
quately if equilibrium and/or continuing existence of the system 
is to be maintained. (ES 16) 

The four problems are those of: 

G Goal attainment 

A Adaptation 

I Integration 

L Latency 

Goal attainment marks the termination of any action cycle. An 
action cycle of a sub-system terminates when it has completed 
its contribution to the functioning of a larger system, that is, when 
the sub-system has attained a goal that is "intermediate" from the 
larger system standpoint. By definition, the goals of any "final" 
system, for example, of society, do not contribute to the function- 
ing of a still larger system, and may therefore be called "ultimate." 
This distinction is not made by Parsons, and he does not use these 
terms, but they appear to be necessary to understanding. The goal- 
attainment problem is that of keeping the action system moving 
steadily toward its goals. 

114 • Chandler Morse 

Adaptation is the process of mobilizing the technical means 
required for (a) goal attainment, and (b) latency. It involves the 
process of inference, which is the heart of what we mean by 
rationality. The adaptive problem is that of properly perceiving and 
rationally manipulating the object world for the attainment of ends. 

Parsons considers that all ends are "goals," and therefore that 
adaptation is relevant only to "goal-attainment." But it is clear 
that, in the Parsonian conception, the maintenance of cooperation 
must be an "end" of some social processes (and a function of 
some roles), though not a "goal" in the sense in which this is 
defined as the termination of an action cycle. Ends imply means, 
and the mobilization of means is the adaptive function. 

Integration is the process of achieving and maintaining appro- 
priate emotional and social relations (a) among those directly 
cooperating in a goal-attainment process, and (b) in a system of 
action viewed as a continuing entity. The integrative problem is 
that of holding cooperating units in line, of creating and maintain- 
ing "solidarity," despite the emotional strains involved in the proc- 
esses of goal attainment and the manner of sharing the fruits 
of cooperation. 

Latency is an interlude between successive goal-attainment proc- 
esses. It is not a period of inactivity; but the activities, whatever 
they may be, consist of restoring, maintaining, or creating the 
energies, motives, and values of the cooperating units and so do 
not explicitly advance the larger system toward its goal. A family's 
home activities are "latent" from the point of view of society even 
though the members of the family may be cooperating in planting 
a garden, which is a goal-attainment process from the family 
standpoint. The latency problem is to make sure that units have 
the time and the facilities, within a suitable conditioning environ- 
ment, to constitute or reconstitute the capacities needed by the 

The four functional problems fit together in the following ways. 
The G- and A-problems taken jointly constitute the "task-orientation 
area" of "instrumental activity" and the I- and L-problems jointly 
constitute the "social emotional area" of "expressive activity." These 
are the two problem areas referred to at the beginning of Section 
I. But, as noted there, the two problem areas are not neatly distinct. 


They merge, though in diflFerent ways, in the two basic strands 
of action process: "task-performance" and "system-maintenance." 
Task-performance heads up in the goal-attainment problem, but 
the process involves a blend of instrumental activity and expressive 
activity. Thus, there are 

. . . two sets of exigencies of goal-attainment, . . . the "exigencies 
of the task-orientation area" and "exigencies of the social emo- 
tional area" of a system of action. More specifically, they are 
the exigencies of adaptation and of integration. (WP 210) 

In other words, task-performance requires solution of both the 
adaptive and the integrative problems as preconditions for goal- 
attainment; the former is always necessary and may be sufficient; 
the latter may be necessary and is never sufficient. 

System-maintenance, one would like to think, should bear a more 
or less symmetrical relationship to task-performance, but Parsons 
is not explicit. The requirements of symmetry would be met if 
system -maintenance headed up in Latency: then we could say 
that the task-performance process terminates in Goal-attainment, 
and that the system-maintenance process terminates in Pattern 
Maintenance and Tension Management— the other term for Latency. 
(ES 17, 19) 

This interpretation is consistent with Parsons' view that goal- 
attainment and latency "designate antithetical, i.e., independent 
directions oi the disposal of the inflow of motivational energy into 
the system." (WP 190) Task-performance directs energies toward 
the attainment of a goal state, and termination of the action cycle 
per se; system-maintenance then directs energies toward attainment 
of a state of latency, in preparation for beginning a new action 
cycle.* The processes involved in both instances are those of 
adaptation ( instrumental activity ) and integration ( expressive activ- 
ity). The relation of adaptation and integration to system-mainte- 
nance is the precise inverse of their relation to task-performance. 

The interdependence between task-performance and system-main- 
tenance may be summarized in the following way: 

Attainment of ultimate system goals is a necessary condition for 

* The term "then" implies logical, not necessarily temporal, sequence. 

116 • Chandler Morse 

meeting unit needs, for conducting and enjoying the activities of 
Latency. But latent interludes, in which system business is in 
abeyance, are the ultimate justification for submitting to the disci- 
pline required by social goal-seeking activity. 

The Four Functional Imperatives Examined More Closely. The 
four functional imperatives, or problems, operate at both a micro- 
analytic and a macro-analytic level in the Parsonian model. (WP 
193, 212) At the micro-level* they purport to specify the phases 
through which individual actors in a small action system and the 
action system as a whole must progress during an action cycle. 
At the macro-level the imperatives provide a means of ( a ) allocat- 
ing roles analytically among four functional sub-systems of any 
given system, and of (b) sorting out the input-output flows among 
these sub-systems. 

An action cycle begins with a commitment of motivational energy 
to a particular concrete goal by one or more hitherto inactive 
member units, the goal being one to which all actors who are to 
participate in the cycle must, through their partially overlapping 
need-disposition patterns, already have a shared set of motivational 
commitments. The performance-sanction interchanges among mem- 
ber units proceed through various functional phases, including 
especially the adaptive (instrumental) phase, until the system has 
both completed the task (goal attainment) and met requisite 
system-maintenance needs (latency), after which the units are 
ready to begin another action cycle. From the system point of 
view, micro-process in a subsystem is a transformation process. 

The Imperatives of the Task Orientation Area: Goal Attainment 
and Adaptation. 

Goal attainment iavolves intrinsically gratifying activity. It is 
the culminating phase of a sequence of preparatory activities. 
(WP 184) 

* The micro-analysis of Working Papers is concerned with small 
group behavior under experimental conditions where a task is as- 
signed to the group and its process of organizing to perform the task 
is observed and analyzed. Much of what goes on is therefore a process 
of "role creation" rather than a process of role behavior. The implicit 
assumption seems to be that there is little essential difference between 
these processes. 


In the consummatory phase the relation to the object is an 
"intrinsically" gratifying (or deprivational) one; in the instm- 
mental phase, on the other hand, there is greater or less distance 
from such a goal-state and activity is directed to altering the 
system-object relationship in the goal-state direction. (WP 211) 

Read Hterally, these statements suggest that all goal states must 
be of an "ultimate" character, meaning that they must be ends in 
themselves, like eating dinner or seeing a play, and not instru- 
mental means to the attainment of some further end— one's own 
or someone else's— like washing dishes or putting on a Broadway 
production. This will hardly do. By far the greater part of all 
human activity is instrumental in the sense (implied above) of not 
providing ultimate gratification, yet we must suppose that most 
of these activities are also "intrinsically gratifying" in some sense, 
as well as instrumental to eventual attainment of an ultimate goal. 
Moreover, we must suppose that Parsons intends to permit this 
construction, for otherwise, contrary to the views put forward in 
Economy and Society and the Administrative Science Quai'teiiy 
article, no factory in the land could have a goal, a goal-attainment 
sub-system, or a goal-attainment phase. 

There is a special significance to the difference between inter- 
mediate social gratifications, which are a necessary condition for 
system functioning, and ultimate personal gratifications, which are 
the final justification of all social activity. The latter are their own 
reward, but the former, no matter how incidentally pleasurable they 
may be to the actor, must be rewarded by the social system. Con- 
sequently there is a sense in which an action cycle is not complete 
until the performances contributory to attainment of intermediate 
goals have been rewarded. That is, social gratification is not enough; 
there must be a matching flow of personal gratifications. But in a 
well-integrated society the latter flow can vary rather widely in 
the short run relative to the former. In other words, there are com- 
mitments to the system that hold it together even when it fails for 
a time to perform satisfactorily. 

The Imperatives of the Social Emotional Area: Integration. A 
system of action would be a somewhat delicate and tenuous struc- 
ture if it were held together only by bonds of common perception, 
understanding, and expectation like those described in Section 11. 

118 • Chandler Morse 

These bonds, whether weak or strong, are continually being under- 
mined by conflicting individual interests, inadequate communica- 
tion, changes in the object world and in the culture system that 
necessarily affect different parts of the action system differently, 
and so on. 

The identity (or integrity) of a system of action is embodied 
in the sense of solidarity that binds its members together, that 
gives them a sense of collective belonging, of mutual interdepend- 
ence, so that they do not require an explicit quid for every quo 
but are prepared to accept a diffuse assurance of the general 
benefits of membership and to make their contributions accord- 
ingly. The family is the prime example of a solidary action system, 
but even the workman who gives his eight hours a day for five 
days a week in return for the general assurance of a pay envelope 
on Friday is showing a degree of solidarity with his employer; and 
the fact that he is willing to accept money, which in itself is value- 
less, is evidence of a solidary relationship to the money-issuing 
authority and the society from which that authority derives. Gener- 
ally speaking, solidarity is always limited. As a rule, no one will 
wholly sacrifice self for all, contintinually "give all," "get nothing." 

Willingness to make specific contributions in exchange for some- 
what diffuse benefits (or even none except "glory" or "reputation") 
is only one major aspect of solidarity. Another is a willingness to 
contribute to maintaining the integrity of the system, in fact, an 
acceptance of responsibility for doing so. Thus it is that much of 
what goes on in a system of action is concerned with integration. 
Integration, it should be re-emphasized, is a necessary aspect of 
both task-performance and system-maintenance processes. The first 
kind of integration is needed because the adaptive process of goal 
attainment can never work perfectly. The need for the second kind 
of integration arises because there is a problem which 

... is closely analogous with that on the adaptive side. Given 
the expectation pattern of one member unit, there is no guar- 
antee that the relations to other units on which the fulfillment of 
the expectation depends will "stay put." There will, therefore, 
be a necessity for processes of adjustment, either by positively 
controlling the relevant unit or by accommodation to it. (WP 


Latency (or Pattern Maintenance and Tension Management). 
Latency consists of two related problems. One— pattern mainte- 
nance—is the problem of stabilizing a set of (latent) commitments 
to a set of goals that has been "legitimized" by the cultural value 
pattern of the system; the other— tension management— is that of 
eliminating the residual "tensions" that occur within member units 
as the result of the fact that no goal attainment process carried out 
by any action system is likely to gratify every participating member 
unit completely. This leaves the "frustrated" units with a problem. 

The two aspects of the Latency problem have as a common 
element the fact that they 

. . . focus on the unit of the system, not the system itself. Inte- 
gration is the problem of inteninit relationships, pattern mainte- 
nance of intraunit states and processes. (ES 50) 

Thus, if a unit's expectation of gratification is not fulfilled, and in 
consequence the unit develops a state of "tension," this is the unit's 
problem, not that of the system. 

Whether or not a given unit maintains or changes its structure 
of commitments to a given set of goal expectations is the unit's 
problem. But stability of system operation requires stability of unit 
operation. When, from the point of view of the larger system, a 
member unit is engaged in mending its fences, the larger system 
is said to be in a state of "latencv." 

When Parsons says that goal gratification and latency designate 
antithetical directions of the disposal of the flow of motivational 
energy into the system he apparently means that in the motivated 
pursuit of a goal the participating units must allocate energy and 
act on the operative assumption that their need-dispositions will 
be gratified ( whatever mental reservations they may hold ) ; whereas 
in the latency phase tlie results of the goal-attainment process are 
an accomplished fact, and motivational energy goes into sorting 
out the consequences for the units and equilibrating their positions. 

Significance of the Functional Imperatives. To understand Par- 
sons it is essential to recognize his aim, which is to establish a link 
between the psychological make-up of the human animal and the 
way in which he organizes social relationships and behavior. That 

120 • Chandler Morse 

the former is in large degree explanatory of the latter is the most 
fundamental premise of Parsons' thought. 

The functional imperatives are a device for moving between indi- 
vidual psychology and social behavior. Man is viewed as purposive 
—hence goal-seeking; he is regarded as rational— hence problem- 
solving: adaptive with respect to the environment and integra- 
tive with respect to his social (and emotional) relationships to 
others of his kind; and he is regarded as an individual, with pri- 
vate, yet socially conditioned, needs— hence needing release from 
the strains of coordinated (competitive, collaborative, or coerced) 
behavior, while ever mindful that even his private life "belongs" 
to society in a sense, and must be led in accordance with certain 
rules. These crucial facts about the human animal, says Parsons in 
e£Fect, explain the "shape" ( or pattern ) of culture— the accumulated 
distillation of human experience; they also explain much about the 
shapes of personality systems and social systems, for these are 
formed by the processes that transmit culture from generation to 
generation, molding the need dispositions of individuals and the 
structure of role responsibilities of society. 

From this Parsons concludes that one should be able to identify, 
in the macro-structure of society, elements that reflect the influence 
of the functional imperatives. Stated otherwise, human experience 
cumulates not only as culture but also as social organization; just 
as ideas become refined and differentiated, so do role responsibili- 
ties. Each process is subject to a high degree of historical variabil- 
ity, and the degree to which social organization has developed 
(become differentiated) is far greater in some societies than in 
others. But these variations should not be permitted to obscure the 
common theme. 

The jump from small group analysis (in Working Papers), via 
family analysis (in Family) and organizational analysis (in the 
Administrative Science Quarterlij article), to societal analysis is a 
large one, and Parsons has taken it with characteristic disregard of 
troublesome detail. In the second part of this paper I shall en- 
deavor to provide some indication of the part played by the func- 
tional imperatives in Parsonian analysis as set forth in Economy 
and Society. 



Macro-structure and Macro-process 

Society as a Social System. In micro-analysis, role specialization 
is viewed in terms of the norms appropriate to performance and 
sanction (WP 202-208) and to role differentiation. (WP 245-254) 
This shades over into macro-analysis, as when the occupational 
subsystem is identified as that group of roles, either individual or 
collective, that is "differentiated from other subsystems of the so- 
ciety by primacy of adaptive functions," * and when sub-groups of 
roles are classified according to their input-output relations with 
each other, or according to the scope of their responsibility. (WP 
254-264) Micro-analysis is concerned with interaction between 
individual role incumbents; macro-analysis with the flows— inputs 
and outputs— among groups of occupants of similar or closely re- 
lated roles. 

No attempt is made in Working Papers to analyze structure and 
process in terms of the four functional imperatives at the societal 
level. This is first undertaken in Economy and Society, which pre- 
sents an outline sketch of a model for society as a whole, and a 
detailed analysis of its economic sub-system. Underlying this model 
is an implicit hypothesis that, as societies become more complex, 
groups of interdependent roles emerge that are specialized in the 
performance of one of the four imperative functions for society. 
In primitive and peasant societies occupational responsibilities are 
assigned to one or both parental roles. In a modern, Western so- 
ciety the typical male head of a family occupies not only the father 
role but also a distinct occupational role, with responsibilities de- 
fined in terms of societal, not family, "ends" and requiring ex- 
tensive interaction with members of society outside the family; 
in addition, he may occupy any number of nonoccupational com- 
munity roles, which ordinarily provide noneconomic (expressive 
rather than instrumental) rewards. 


* Later modified. In Economy and Society some occupational roles 
have primacy of other functions (G, I, or L). 

122 • Chandler Morse 

The thesis of Economy and Society may be epitomized as fol- 

a. Every society, being a social system, must contain roles with 
responsibilities for solving the four basic system problems at the 
societal level. This is the theme of Chapter I. 

b. When the scale and complexity of these problems becomes 
sufficiently great, there is a "division of labor," and roles appear 
that have primary responsibility to contribute to solution of only 
one of these problems. Consequently, any sufficiently complex so- 
ciety will be found to have four sets of specialized roles, one for 
each of the basic system problems. Each such set of specialized 
roles, it is contended, constitutes a sub-system of society and obeys 
the laws governing the operation of social systems. (ES 13-19, 53) 
These sub-systems are designated as follows: 

G The PoHty 
A The Economy 
I The Integrative sub-system 

L The Latency, or Pattern-maintenance and Tension- 
management, sub-system 

Not all the sub-systems need be developed ( diflFerentiated ) to the 
same degree, and at any particular historical stage of development 
they presumably will not be.*^ It will be noted, for example, that 
the I- and L-sub-systems have no distinctive names. This suggests 
that they are less highly diflFerentiated than the other two— that 
social evolution has not proceeded far enough for these functions 
to be recognized in the language. But Parsons appears not to rec- 
ognize the possibility of varying degrees (stages) of diflFerentiation. 
He implies that the United States is fully diflFerentiated, and that 
the same is true of its sub-systems. This seems unlikely, 

c. Within these broad lines of specialization there can be an 
indefinitely large number of sub-specializations, depending on the 

* An historical analysis of the process of structural differentiation will 
be found in Neil J. Smelser, Social Change in the Industrial Revolu- 
tion (Chicago: University of Chicago Press, 1959). Smelser makes 
use of the seven "levels of generality of economic resources" outlined 
in Economy and Society, pp. 138-143. Limitations of space preclude 
discussion here of this potentially useful tool of analysis. 


technical requirements arising from the scale and complexity of 
development, and the state of knowledge. In Chapter II of Econ- 
omy and Society the fourfold breakdown is applied to the Econ- 
omy, and each of its sub-systems is subdivided in the same way. 

d. The terms on which interactions ( performances and sanctions ) 
occur among roles within one of the four sub-systems difiFer in sig- 
nificant ways from those governing interactions between roles in 
different sub-systems. In the first case, both parties recognize that 
in some sense they belong to a common social system, share re- 
sponsibility for attaining a common goal (which is that of their 
specialization), and therefore stand in a "solidary" relationship to 
each other. This implies that the two role occupants also share a 
responsibility to maintain social order in the swZ?-system. In the 
second case, any solidarity between the parties must relate to the 
next higher system level— that of society as a whole— and their 
shared sense of responsibility for system maintenance includes 
only an ill-defined obligation to maintain order in the respective 
sub-systems. Since the sense of solidarity is more dilute at higher 
system levels (unless strongly reinforced by a large input of in- 
tegration, as in time of war), and the requirements of system 
integrity are less clearly defined, performance-sanction interchanges 
across sub-system boundaries permit a degree of mutual disregard 
of consequences for the stability of the opposite sub-system. There 
is thus an element of antagonism and conflict in transactions across 
sub-system boundaries. 

e. To prevent these antagonisms from getting out of hand certain 
control mechanisms— regulative institutions, and so on (SS 137-150) 
—must come into existence. The institutional structures of markets 
perform this function in the Economy. (ES ch. Ill) 

f. Recognition of the functional role of the Economy in society, 
and of its own internal differentiation along functional lines, will 
lead to a better understanding of economic processes and of the 
processes of institutional change. (ES ch. IV, V) 

We cannot, in the space available, do justice to each of the pre- 
ceding contentions. Nonetheless, we shall endeavor to indicate the 
nature of each of the four sub-systems when viewed in societal 
terms; to say something about the Persons' views concerning the 
organization of the Economy, including its most distinctive institu- 

124 • Chandler Morse 

tions; and to see what insights, if any, this elaborate analysis may 
contribute to an understanding of economic process. 

The Macro -structure of Society. ^ 

The A-sub-system: The Economy. The Economy, which is 
specialized in the adaptive function of society, is regarded as pro- 
ducing generalized facilities as means to an indefinite number of 
possible uses, these facilities being wealth and income. ( ES 47-8; 21 ) 

Negatively . . . [this adaptive function] implies the minimization 
of subjection to control by the exigencies of the external situa- 
tion (e.g., floods, famines, shortages, etc.). Positively it implies 
the possession of a maximum of fluid disposable resources as 
means to attain any goals valued by the system or its sub-units. 
The general concept for these disposable resources is wealth 
from a static point of view and income from the point of view 
of rate of flow. (ES 21) 

The Adaptive function is necessarily linked closely to the Goal- 
attainment function, since adaptation is concerned with "the prob- 
lem of controlling the environment for purposes of attaining goal 
states." In fact, 

. . . when a social system has only a simply defined goal, the 
provision of facilities, or the "adaptive" function, is simply an 
undifferentiated aspect of the process of goal attainment. (ES 
18; italics added) 

But in complex systems, with a plurality of goals, the distinction 
between the Adaptive and the Goal-attainment functions is sharp- 
ened because it is necessary to pursue many goals that are inter- 
mediate; and these goals, from the point of view of society, are 
goals of its Adaptive sub-system only. At the societal level, there- 
fore, the "differentiation between goal attainment and adaptive 
processes is often very clear." (ES 18) 

To clarify the distinctions made in the preceding quotations, 
consider the proprietor of a small restaurant who does all the work 
himself. The goal of this little system is to provide meals for cus- 
tomers, and the proprietor's responsibility to see that the goal- 
attainment function is discharged requires him to make all the 
decisions concerning what to serve, where and what to buy, what 


price to charge, and so on. His responsibility for the adaptive 
function requires him to prepare the food, serve it, pay bills, and 
so on. These activities, though distinct in kind, are "an undiffer- 
entiated aspect of the process of goal attainment" because the G- 
and the A-responsibilities are embodied in a single role. But if the 
restaurant comes to employ a chef, waiters, cashier, bookkeeper, 
all of whom will be engaged in performing the adaptive function, 
leaving the proprietor free to concentrate on seeing that the job is 
well done, the differentiation between the adaptive and the goal 
attainment processes through differentiation of roles becomes quite 

The Economy is not identical with what we mean by "business." 
All concrete units, and not only business firms, participate in the 
Economy (ES 14). Moreover, business firms are concrete organiza- 
tions, each of which contributes to solution of all four system prob- 
lems, though primarily to the adaptive. That is, business firms and 
other concrete organizations (and roles) are seldom differentiated 
perfectly. Hence business firms appear analytically in all four of 
the primary functional sub-systems. 

The G-sub-system: The Polity. A basic distinction is drawn be- 
tween the production of wealth and income and their actual use 
for the attainment of system goals. This seems to mean that there 
is conceived to be a basic distinction between ( 1 ) the allocation of 
resources and (2) the distribution of income. Economic theory 
treats these as two aspects of a single process. When Parsons im- 
plies that the former is the function of the Economy, the latter 
of the Polity, he is therefore making a sharp but perhaps important 
break with a well-established intellectual position. 

Wealth is one, but only one, of the indispensable prerequisites 
needed for the Polity to perform its function. 

To put it in a slightly different way, the goal of the polity is to 
maximize the capacity of the society to attain its system goals, 
i.e., collective goals. We define this capacity as power as dis- 
tinguished from wealth . . . [though wealth is] an ingredient 
of power. . . . (ES 48) 

Power is the generalized capacity to mobilize the resources of 
the society, including wealth and other ingredients such as loyal- 
ties, "political responsibility," etc., to attain particular and more 

126 • Chandler Morse 

or less immediate collective goals of the system. (ES 49; italics 

Thus the goal of the Polity is to maximize power. 

That is why Parsons places the system of banking and finance, 
which exercises a measure of control over purchasing power, in the 
Polity. It is also why he says: 

Contrary to much previous opinion, we feel that "classical capi- 
talism," characterized by the dominance of the role of owner- 
ship in the productive process, is not a case of full "emancipa- 
tion" of the economy from "political" control, but rather a 
particular mode of such control. This follows from our view 
that ownership is anchored essentially in the polity. (ES 285-6) 

Remember Marx? "The State is the executive committee of the 

The I-sub-system. The integrative sub-system of society 

... is the "producer" of another generalized capacity to control 
behavior analogous to wealth and power . . . [i.e.,] "solidarity." 
. . . Solidarity is the generalized capacity of agencies in the 
society to "bring into line" the behavior of system units in ac- 
cordance with the integrative needs of the system, to check or 
reverse disruptive tendencies to deviant behavior, and to pro- 
mote the conditions of harmonious cooperation. (ES 49) 

It is, of course, possible that the adaptive activities of the mem- 
ber units of an action system "may be mutually supportive and 
hence beneficial to the functioning of the system." But it is also 
possible, and indeed likely to a degree, that they "may be mutually 
obstructive and conflictful." (ES 18) Hence it becomes necessary 
to "produce" a certain amount of solidarity by taking steps to re- 
ward good performance by positive sanctions (rewards), to punish 
poor performance by negative sanctions (penalties), and in general 
to make the member units feel that their activities have been prop- 
erly "appreciated." 

The L-sub-system. Performance of the Latency function con- 
tributes to system-maintenance. Especially it involves determina- 
tion of the extent to which the actual goal-attainment consequences 
of any action process have conformed to the norms (ideal expecta- 



tions) specified or legitimized by the value system. The interaction 
of member units performing latent roles is not part of the action 
cycle of the system to which the Latency sub-system belongs. The 
units are not pursuing system goals; they are merely restoring 
themselves and each other to normal functioning states, both as 
biological and psychic organisms and as the properly socialized 
agents, so to speak, of future action cycles. 

The function of the Latency sub-system is to contribute stability 
to the institutionalized norms and internalized motivational com- 
mitments that constitute the basic structural elements of action 
systems. These norms and commitments, says Parsons, tend to 
change under pressure of influences arising (a) in the culture sys- 
tem and (b) as the result of "abnormal" motivational tensions 

... arising from "strains" in any part of the social situation or 
from organic or other intra-personal sources . . . [which] threaten 
individual motivation to conformity with institutionalized role 
expectations. (ES 17) 

Stabilization against destabilizing influences arising in the culture 
system is "pattern-maintenance"; that against similar influences 
arising from strains is "tension-management." They 

. . . difi^er from the integrative problem in the sense that they 
focus on the unit of the system, not the system itself . , . [hence 
latent, which implies that] essential conditions of the larger 
functioning, rather than the functioning itself, are involved. 

(ES 50) 

Structure and Process in the Economy 

The flows among Parsons' four societal sub-systems are shown in 
Figure 1. The G-sectors * of the A- and L-sub-systems are set in 
juxtaposition to each other, as are those of the G- and I-sub-systems. 
Between A and G and L and I, the flows are handled bv the 
A-sectors of the sub-systems. Two of the other pairs of flows are 
handled by the G-sectors and two by the I-sectors. No justification 

* The term "sector" is employed here in lieu of the impossible "sub- 

128 • Chandler Morse 

other than that of diagrammatic symmetry is offered for these 

Another feature is that there is no L-sector and no L-boundary 
in any of the sub-systems. The following rather cryptic explanation 
is reproduced without essential omission: 

The latency subsystem of any larger system is always a special 
case relative to the other three systems in the sense that it is 
"insulated" from sensitivity to the current performance-sanction 
interplay of the larger system with its cognate systems. To be 
sure, the latency subsystem of the society has boundary rela- 
tions with the . . . nonlatent systems. . . . But its own latency 
subsystem is not contiguous to any subsystem of any other 
primary system. . . . [This] "special" boundary of the latency 
subsystem at any given system level is a cultural rather than an 
interaction boundary. The latency subsystem . . . maintains value 
patterns . . . [which are] not interactive. (ES 69) 

Figure 1 enables us to identify what Parsons regards as the 
functional output of each sub-system, since this output goes out 
across the G-boundary in each case. The goal of the Economy thus 
turns out to be the production of consumer goods and services, 
which is considerably narrower than the production of "wealth and 
income" referred to earlier. The explanation is that the production 
and disposition of capital goods is conceived to be a circuit within 
the Economy, and not to involve an output to the rest of society. 
This implies that production of capital goods is an intermediate 
goal of a sector of the Economy which is not shown on the diagram. 
The goal (social function) of the L-sub-system is the provision of 
labor services to the Economy. The goal of the Polity is "imperative 
coordination," which is unexplained; and the goal of the I-sub- 
system is "contingent support," also unexplained. 

The several output flows shown in Figure 1 are the macro- 
analogue of performance in the role expectations sense of that 
term. There must therefore be balancing flows, analogous to sanc- 
tions, in the opposing directions. These are provided by what 
Parsons calls the "double interchanges at the boundaries." These 
double interchanges between sub-systems are shown completely 
only for the Economy ( Figures 2, 3, 4 ) . 




THE ECONOMY Sul>- system 

G / 

Sub-system Sub-system 



































Sub- system THE POLITY 





Sub- system 

Sub- system 


Sub- system 

(Aspects of 








Sub- system 



Sub- system 

*We may abstract from this 
diagram the general schematic 
representation for the primary 
boundary interchanges within 
any system as follows: 

Figure 1. Boundary interchanges between the primary sub-systems 
of a society. (From Economy and Society, p. 68.) 


Chandler Morse 

To economists, Figure 2 has a familiar yet incomplete appear- 
ance. It would be a perfect "circular flow" model for a peculiar 
society that had only households as buyers of goods, only labor as 
a factor of production, and only consumer goods and services as 
output. The reason for what seems at first to be a flagrant flouting 
of the economic facts of life is that the factors of production other 
than labor, and the distributive shares other than wages, are found, 
explicitly or implicitly, in the other figures. Investigation shows that 
they are all accounted for and that, although the Parsonian circular 
flow differs from the economic, the two can be reconciled. 


"Economic" Decisions 

Decision to 

offer employment -<- 

Decision to 

Labor Services 


Consumer Spending 

Consumer Goods 

and Services 

** Household" Decisions 

Decision to accept 

Decision to 
"*~ purchase 

Figure 2. The double interchange between the economy and the 
pattern-maintenance sub-system. (From Economy and 
Society, p. 71.) 

Consider the factors of production and the corresponding dis- 
tributive shares. Labor and wages are accounted for in Figure 2. 
Land, a somewhat (but not unreasonably) broadened concept, is 
said to be a contribution across the "special" Latency boundary 
of the Economy, which appears in none of the figures. In Parsons' 

. . . land is a specific instance of a pattern-maintenance factor. 
The common denominator of the three "land" categories— physi- 
cal facilities, cultural facilities, and motivational commitments- 
is a certain order of control to which they are subject. They are 
committed to economic production on bases other than the 
operation of short-term economic sanctions. They are "fed into" 



the economic machine prior to current operations; consequently 
they must be treated as a given determinant of subsequent 
processes. (ES 70) 

Capital as a factor of production does not appear in the double 
interchanges. Instead, there is an item "control over capital funds" 
(Figure 3). Capital goods, as indicated earlier, are retained within 


"Economic" Decisions 

Decisions to borrow or 
otherwise obtain liquid re- 

Control over 
Capital Funds 

"Political" Decisions 

Decisions to supply liquid 
resources through crea- 
tion of capital funds 

Decisions to capitalize or 
otherwise enhance pro- 
ductivity capacity 

Rights to Intervene 

Encouragement of 
Productive Enterprise 

Control of Productivity 

Decisions to encourage or 
discourage enterprise 

Figure 3. The double interchange between the economy and the 
polity. (From Economy and Society^ p. 77.) 

the Economy for its own further use and are counted, presumably, 
as an addition to wealth. This is a sensible enough procedure, but 
one wonders whether it was intentional. The sanction for use of 
capital funds is shown in Figure 3 as "rights to intervene," and it 
appears from textual discussion (ES 27, 75-76) that this category 
is meant to provide for the payment of interest. 

Finally there is "entrepreneurship" which, since the time of Al- 
fred Marshall, has been generally admitted to the circle of classical 

132 • Chandler Morse 

factors o£ production, and its sanction, profit. These are found in 
Figure 4. 

"Economic" Decisions "Integrative" Decisions 

Enterpreneurial Service 

Decision to offer oppor- Decision to offer inte- 

tunity to entrepreneurs grative services to the 



Demand for new 

-« — — 

product dombinations 

Decision to innovate Decision to change con- 

sumption patterns 
New output combinations 

Figure 4. The double interchange between the economy and the 
integrative sub-system. (From Economy and Society, 
p. 79.) 

It thus appears that Parsons has provided a thorough roundup 
of economic input, output, and income elements after all. The 
major diflFerences between the Parsonian and the standard econo- 
mic macro-models are: (1) Parsons' handling of expenditure, which 
is distributed over a number of sectors, and "rent," which has no 
specified recipient; (2) the fact that Parsons excludes certain oc- 
cupations from the Economy, thus having, in effect, a category of 
"unproductive" labor; * and (3) the fact that there is no explicit 
provision for saving or for the return of unsaved rent, interest, 
profits, and "unproductive" wages to the circular flow via pur- 
chases of consumer goods and services. We cannot be sure whether 
the filling of these gaps would violate any crucial Parsonial prem- 
ises, but it seems unlikely. Parsons also includes certain symbolic 

As did Smith and Marx, though their classifications differed. 


flows— rights to intervene and the Uke— that do not appear in stand- 
ard economic models. All of the postulated flows between the 
Economy and the other sub-systems, together with those that are 
internal to the Economy, are shown in Figure 5. 

Institutional Structure of the Economy. 

Contract. The flows that occur within the Economy and be- 
tween the Economy and the other sub-systems are visualized as 
constituting aggregates of transactions (exchanges). In general, 
each flow involves the exchange of an "input" (factor of produc- 
tion) for an "inducement." The process of reaching a contractual 
settlement of terms— designated as "contract"— is seen to involve, 

. . . first, the process of bargaining for advantage, in which each 
party, with particular goals and interests and the particular ad- 
vantages or disadvantages of his position, seeks to make the best 
possible bargain; second, the socially prescribed and sanctioned 
rules to which such bargaining processes are subject, such as 
the guarantees of interest of third parties, restrictions on fraud 
and coercion, and the like. (ES 104-105) 

The institutional structure of the Economy— the institution of 
contract— is concerned with the second of these two sets of de- 
terminants; but the outcome of exchange processes will be strongly 
influenced by the first of them. 

To see what elements of contract are likely to become insti- 
tutionalized it is necessary to recognize that a transaction links two 
distinct systems of action; that the behavior of the transactors must 
be articulated and regulated; and therefore that the two behavior 
patterns must be so integrated that they constitute a "partially 
independent social system." (ES 108) The four functional problems 
of this system are identified as: 

G The arrival at mutually advantageous terms for the exchange 
of primary input for primary output; that is, reciprocal goal- 
attainment based on pursuit of self-interest is the goal of a 
market viewed as a social system. 

A Adaptation to the limiting conditions (determinants of bar- 
gaining power), such as the type of firm represented by a 
salesman and the kinds of goods he sells; the income of a 






























136 • Chandler Morse 

household and its standards of taste; and the interests of those 
(the firm and the family) who are represented by the re- 
spective transactors. 

I Recognition of the legitimacy of each transactor's goals and of 
his normal competitive role behavior; such recognition is rep- 
resented by the symbolic value of the "secondary" perform- 
ances and sanctions that enter into every exchange: approval, 
esteem, success, for example. 

L The common value pattern, or mutual recognition of the con- 
tribution of each transactor to attainment of shared goals: for 
example, a common valuation of the social function of pro- 
duction, and therefore of all the transactions involved in it. 

Markets. Contract finds concrete expression in the institutional 
characteristics of markets, which vary according to the nature of 
their constitutive transactions. One of the central problems of 
modern economic theory has been to relate market behavior to 
market structure. The benchmark type is the "perfect market," in 
which no seller or buyer can influence price. In addition, a great 
variety of imperfect markets has been recognized to exist. These 
have been classified according to the particular conditions under 
which sellers or buyers can influence price, each such set of con- 
ditions yielding a different pattern of rational behavior. In economic 
theory, the bases of distinction among market structures have been, 
principally: the number of sellers or buyers, differentiation of prod- 
ucts, ease of entry or exit, and "cross-elasticities" of demand— the 
effect of a change in the price of one seller's product on the demand 
for that of another. 

When Parsons says that "certain economic theories of imperfect 
competition concern the G and partially the A components of the 
market relationship" (ES 144), and that "there has been a nearly 
complete lack of attention to the I- and L-components" (ES 146), 
he is referring to these traditional criteria of market structure. His 
own approach takes a different line. A perfect market is defined 
as one in which there is "(1) either sufficient regulation or sufficient 
competition so that the settlement of terms is not skewed toward 
the advantage of either side," which implies an "equality of power" 


and "(2) symmetry of 'type of interest'" with respect to all the 
components except G ("which is the focus of the power factor.") 
"On these grounds, plus the usual economic ones, only a market 
internal to the economy can approximate the ideal type of market 
perfection closely." (ES 146) 

The most noteworthy aspect of this approach is its direct rejec- 
tion of the initial neo-classical economic view of market perfection. 
In that conception, the inability of any seller or buyer to influence 
price was held to imply that the resulting price, which could not be 
skewed to the advantage of any market participant, would be "op- 
timal"; and it was usually inferred that if every market in a society 
was perfect in this sense, and therefore productive of market opti- 
mality, the aggregate result would be socially optimal. This, it came 
to be realized, would have to mean that the price structure of so- 
ciety could not be skewed to the advantage of either sellers as a 
whole vis-a-vis buyers, or of buyers vis-a-vis sellers. Critics of this 
view observed that diflFerences in income distribution ( which would 
give some buyers more influence than others) would skew results 
even under universal "perfect competition," but this has been about 
as far as dissension has gone among economic theorists.* While not 
denying the relevance of this limited criticism, Parsons contends 
that skewing necessarily results from the fact that initial control 
of the factors of production lies in the noneconomic sub-systems 
of society. (ES 104) In the exchange of factors for products be- 
tween these sub-systems and the Economy, therefore, the criterion 
of "symmetry of interest" is violated. Consequently, such exchanges 
will be considerably influenced on one side of the market, but not 
on the other, by various noneconomic interests and values. Con- 
trary to the basic tenet of orthodox economic theory, such ex- 
changes cannot be governed entirely by the principles of "eco- 
nomic rationality" on the household side of the market. Economic 
rationality, says Parsons, organizes behavior on the basis of a high 
valuation of production as a goal, whereas other forms of rational- 
ity are relevant for systems that have noneconomic goals. (ES 176) 

* This is curious in view of Adam Smith's discussion of the sociologi- 
cal factors that gave masters as a group a bargaining advantage 
vis-a-vis workers. 

138 • Chandler Morse 

This broad approach to the analysis of markets is employed, with 
much insightful comment, to point up the important distinction 
between the market for personal services and that for property; 
and between the markets for ordinary labor, for executive services, 
and for professional services. In each case, an attempt is made to 
show how the nature of the noneconomic interests of those con- 
trolling the factor a£Fect the terms of exchange. The conclusions 
are frequently no diflFerent from those noted by institutional econo- 
mists, socialists, and other critics of neo-classical theory, but their 
derivation from a general sociological theory lends them special 
interest. A few examples will illustrate.*^ 

Labor unions, it is recognized, have the primary function of cor- 
recting the inequality of bargaining power between employers and 
workers. Business firms regard the contract of employment in eco- 
nomic terms, which means that they would like, if they could, to 
treat labor services as a commodity; the worker, manifestly, cannot 
do so and retain self-respect and confidence. An important, semi- 
ritualistic function of unions therefore has been to integrate the 
worker and his family into a larger collectivity, to lend dignity and 
prestige to worker roles, and to give the worker a basis for con- 
fidence that the operation of the system will take his interests into 
more adequate account. Thus "the union helps the worker to recon- 
cile his inevitable involvements in both firm and household with 
each other." (ES 149) 

A crucial difference between worker and executive is that the 
former has a low and the latter a high level of responsibility in the 
firm. Conversely, the worker's responsibility for the security and 
prestige of his household is paramount, whereas a relatively large 
share of responsibility for the security of the executive's household 
is assumed by the firm and much of that for prestige by the wife 
(because the demands of the husband's job are so pressing). More- 
over, a high level of loyalty to the firm is not a requirement of the 
worker role but it is essential for the executive. Whereas the rela- 
tion between the worker— whose strongest ties are outside the firm 
—and the employer is essentially one of hostility (competition). 

* The four following paragraphs enlarge upon what appears to be 
Parsons' plain meaning. (ES 146-152) 


that between the executive and the firm is one of soHdaritv (col- 
laboration); executives need no "bargaining agency." The payment 
of wages for labor services thus approximates closely to an eco- 
nomic exchange, but the payments of salaries for executive services 
is more in the nature of a transfer, in which the quid and quo are 
not expected by either party to be in precise balance, and in which 
the symbolic aspects of a large salary, a twelve-hour day and seven- 
day week, and a secure position are of great importance. 

The market for professional services has yet a different character. 
The professional is expected above all to be loyal to the ideas, 
ethics, and standards of his profession rather than to the employer. 
In case of conflict of interests, "his freedom to give weight to 
'professional integrity' is at least an implied element in the contract 
of employment." (ES 152) In the civil service and in higher aca- 
demic posts, formal security of tenure protects the professional 
against the danger that this implicit element will not be respected 
in time of tension. In the case of civil servants, loyalty to the public 
interest might transcend that to the party in power; in the case of 
teachers, loyalty to ideas may take precedence over loyalty to a 
particular university (or even a particular form of society). 

Thus, the right of the professional to be rather neutral as regards 
loyalty to his employer contrasts with an expectation of solidarity 
for executives and hostility for workers. In terms of the model, 
workers in the mass, when viewed in the perspective of the Econ- 
omy, take on the appearance of a mere element in the "situation of 
action"— a mass of labor power to be manipulated in the process of 
production. The labor problem arises because this view is in con- 
flict with that of the workers, who need to become integrated mem- 
bers of the firm or the Economy as a social system. The fact that 
this does not always happen, and usually happens imperfectly at 
best, is an aspect of what Marx identified as "alienation," An im- 
portant function of labor unions, then, according to the argument 
advanced earlier, is to de-alienate the worker. Executives, by con- 
trast, are in a position to determine who shall not be treated as 
integrated members of the firm or the Economy, and for what pur- 
poses. That is, executives are in a favorable position to influence 
resolution of the self-collectivity dilemma in the interests of the 
firm (and of themselves). Many professionals are not employed by 

140 • Chandler Morse 

firms but by universities, research agencies, or households, the 
primary function of which is to contribute to solution of the L- 
problem. The fact that, in these cases, neither party to the con- 
tract of employment has a primarily economic (adaptive) function 
has a great deal to do with the important part that noneconomic 
elements play in the professional contract of employment. 

Some Observations on Economic Process. Hoping to demonstrate 
that the fourfold analysis of social structure can contribute di- 
rectly to the solution of theoretical problems in economics, the 
authors of Economy and Society address themselves in Chapter IV 
to business cycle theory. The attempted demonstration fails. Some 
valid observations are made concerning the processes of consumption 
and investment, the contention being that the conceptualization and 
empirical investigation of these processes would be improved if 
sociological determinants were taken into account. Several of these 
determinants are specified, but it is most unlikely that gain would 
result from efi^orts to incorporate them in economic analysis. Family 
spending patterns, for example, are said to involve four categories 
of expenditure, each of which is designed to solve one of the basic 
system problems. Whether or not this be true, economists have 
observed that expenditure patterns of families diflFer according 
to income level (roughly proportional to social status) and that 
hypotheses concerning the form of the consumption function and 
changes in it through time should take this into account. To assert 
that diflFerent spending patterns reflect the differing solutions to 
the four problems appropriate to families occupying different social 
positions seems to add nothing of value for the determination of 
the consumption function. 

The discussion of the investment function is equally sterile. The 
fact that the investment market is relatively unstructured, leaving 
a great deal of room to maneuver, is made the basis for some inter- 
esting observations concerning stock market behavior, but little 
else of consequence is offered. 

A discussion of the structure of the Economy and its internal 
processes precedes the commentary on the business cycle. The 
Economy, like society, is viewed as comprised of four sub-systems, 
as follows (ES 198ff.): 


G Production sub-system 

A Investment-capitalization sub-system 

I Entrepreneurial sub-system 

L Economic commitments sub-system 

Each of these sub-systems, which is further subdivided according 
to the fourfold schema, has interchanges over its "open" boundaries 
with each of the other sub-systems. All this is shown in Figure 5, 
which also shows the "external" interchanges of the Economy with 
the other societal sub-systems. 

The descriptive justification for this elaborate framework is not 
always clear or convincing. Some specific complaints against it 
will be registered in the two concluding sections. 


Economic and Noneconomic Rationalities 
as Determinants of Human Behavior 

That Parsonian theory will stand as a massive landmark in the 
intellectual development of the social sciences is scarcely open to 
doubt; that most of it will undergo fundamental revision is equally 
probable. There is much that is dubious in the theory, and much 
in the exposition that is otiose. Often there are chasms between the 
theoretical abstractions and the concrete reality whose inner mean- 
ing they purport to illuminate. The concepts are strange, with a 
frequent ring of irreality about them, and ambiguities abound. Yet 
these are all remediable faults, provided only that there is a hard 
central core of meaning that can and should be preserved. The 
contention here is that there is such a core of meaning. 

Most major theoretical advances rest on an essentially simple yet 
fundamental insight. For Adam Smith it was the principle of the 
self-regulating economy, achieved through competitive application 
of the rules of economic rationality ( as epitomized in the metaphor 
of the "invisible hand" ) . For Darwin it was the principle of natural 
selection, achieved through a competition conducted according to 
purely biological rules. For Parsons it is the principle that society 

142 • Chandler Morse 

as a whole achieves self-regulation through the application of 
several distinct rationalities— each of which is the cumulative product 
of learning through experience— to the solution of certain peculiarly 
social problems, and through the institutionalization of a vast num- 
ber of behavioral rules which in part embody these rationalities 
and in part complement them. The term "rationality" is used here 
to signify a commonly accepted set of problem-solving procedures 
that are regarded as appropriate in a defined context. (Compare 
the footnote on p. 145 infra.) Thus, economic rationality is appro- 
priate in a market context for deciding whether to raise wages, 
but not in a philanthropic context for deciding whether to increase 
one's gifts to the poor. The rationality of "pattern-maintenance and 
tension-management" is perhaps applicable in the second case. 

Until Parsons, only economics among all the social disciplines 
could be said to have a rational foundation for its theoretical 
formulations. That is, economics alone postulated a set of objective 
rules that could be said to guide, or could be appealed to as appro- 
priate guides for, goal-seeking behavior. That it was able to do so 
was undoubtedly responsible for its theoretical sophistication; its 
relative success in the realm of practical policy may be taken as 
evidence that its rationality postulate has a considerable measure 
of empirical validity. The form and content of the rules for econom- 
ically rational action may vary with the concrete structure of the 
society in which they are to be applied, but there appears to be 
an invariant quality that carries over from one type of society to 
another— for example, from capitalist to socialist societies. 

Parsons has opened the way for other social disciplines to acquire 
distinctive rationalities of this same type. The hypothesis concern- 
ing the rational foundations of action process was stated initially 
in The Structure of Social Action, where it was epitomized in the 
following words: 

The starting point, both historical and logical, is the conception 
of intrinsic rationality of action. This involves the fundamental 
element of "ends," "means" and "conditions" of rational action 
and the norm of the intrinsic means-ends relationship. The ra- 
tionality of action in terms of the latter is measured by the con- 
formity of choice of means, within the conditions of the situation, 
with the expectations derived from a scientific theory (however 


elementary and empirical) applied to the data in question. . . . 
Action in these terms is rational in so far as there is a scientifi- 
cally demonstrable probability that the means employed will, 
within the conditions of the actual situation, bring about or 
maintain the future state of affairs that tlie actor anticipates 
as his end. (SA 698-9) 

The next step (General Theory and Working Papers) was to 
sense the possibility of different rationalities, each differentiated 
according to the character of ends and the appropriateness of 
means; to postulate that there was a minimum number of mutually 
independent objective and subjective "dimensions" (the pattern 
variables) according to which ends and means could be con- 
ceived, perceived, and evaluated; and to postulate further that 
there was a minimum number of basic functional problems— specif- 
ically, four— that had to be "solved" by any social system that was 
to survive. These four functional problems represent four distinct 
(yet interdependent) social "ends," and constitute the basis of four 
corresponding rationalities, the simultaneous application of which 
is responsible for the ways in which social systems function. Within 
a rather undifferentiated social system, such as a primitive family 
or tribe, consistency among the four rationalities and their applica- 
tion is achieved by the institutionalization of role patterns together 
with the opportunity for adjustment by direct settlement of con- 
flicting interests. Notwithstanding, there may be conflicts; they 
are possible even within the personality system! When a society 
becomes highly differentiated, so that the application of each 
rationality becomes a function of specialized roles, the possibilities 
of inconsistency become far more numerous (though not neces- 
sarily more serious). The degree of consistency achieved is a deter- 
minant of the stability or instability of the system, any inconsistency 
among the four types of rationality or their application being a 
particularly important source of conflict and, potentially, of change. 

The pattern variables are properly to be regarded as the "ele- 
mentary particles" from which a plurality of concrete evaluations 
is constructed. Similarly, the four functional problems are to be 
regarded as the (related) elementary particles from which is con- 
structed a plurality of concrete ends; and the corresponding ration- 
alities are the elemental components of all concrete systems of 


144 • Chandler Morse 

action, where "action" is defined as goal-seeking behavior and is 
held to be rational to the degree that the behavior— that is, the 
mobilization of means— is cognitively controlled. 

One is entitled, of course, to doubt that Parsons has correctly 
identified the postulated elements. Further developments may sug- 
gest quite different sets, and there may never be final agreement 
on what the elements are. Yet the hypothesis that there is an 
identifiable set of such elements promises to introduce far more 
rigor than hitherto into the noneconomic social disciplines. It also 
points the way to a much needed modification of economic theory; 
for it permits account to be taken of the limiting effect of non- 
economic rationalities upon the exercise of economic rationality 
without destroying the content of the latter (as some critics would 
do) and so tossing out the baby with the bath. 

One may also question the implied proposition that societies 
must "solve" the four functional problems in order to survive. As 
there is no criterion of "solution" other than the fact of survival 
itself, the proposition implies that mere survival may be taken as 
evidence that a society's value system and social structure are 
"functional." The proposition that solution of the functional prob- 
lems is necessary for survival is stronger than necessary, however. 
It may be replaced by the weaker empirical hypothesis that all 
social action consists of the pursuit of concrete ends which, on 
analysis, will be found to involve attempts to solve all of the 
postulated functional problems, with wide variability in the quality 
of the actual outcome. Survival would thus be consistent with social 
action that yielded exceedingly poor results by any standard of 
evaluation we might choose. The quality of the consequences of 
action, relative to some realistically postulated standard, would 
therefore be a possible alternative to survival as a criterion for 
determining whether or not a value system and social structure 
were functional; and a decline from any achieved level of quality,, 
rather than reduced chances of survival, would be evidence of 

Finally, we must note that Parsons has not attempted to suggest 
a costs -benefits calculus or its equivalent to be applied in the pur- 
suit of noneconomic ends. That is, no empirical procedure com- 
parable to that which lies at the foundation of economic theory 


is proposed for choosing among alternative methods of performing 
the functions of goal attainment, integration, and latency. Nor is 
any rational procedure proposed for setting priorities among the 
four functions when this is rendered necessary by scarcity of 
means. Parsonian theory implies that such rational procedures exist 
but it endows them with no material basis. This is important unfin- 
ished business. It requires that noneconomic values, both negative 
(costs) and positive (benefits), be given an empirical content and 
measurability analogous to that for economic values. The rational 
procedures available to a society for maximizing benefits and 
minimizing costs are determined by its sociocultural level.* With 
knowledge or imputation of such procedures it would become 
possible to determine whether they are consistent with each other, 
to indicate the empirical consequences that would follow from 
their application, and to identify their effects on the quality of a 
society's functioning. 

Questions and Suggestions Concerning 
Structural Differentiation 

A major virtue of the fourfold functional schema, but also a 
source of difficulty, is that it combines elements that are rooted 
firmly in individual psychology ( goal-oriented and adaptive behav- 
ior) with elements that have a peculiarly social and cultural 
significance (integrative behavior and maintenance of the value 
structure). A great deal of The General Theory and The Social 
System is devoted to establishing that the social and cultural 
elements must be and are embodied in the personality, thus estab- 
lishing an essential link among the four elements. But this necessary 
link will also be sufficient only if the existence of collective goals 

* This statement implies that: (a) solutions to social problems in all 
societies, however primitive, reflect the influence of reasoning; (b) 
there is something in the reasoning process— call it logical inference— 
wliich is independent of empirical reference, although different pat- 
terns of inference may exist and are in fact observed; ( c ) the available 
inference procedures, together with the perceivable qualities of ob- 
jects, the behef system, the values to be manipulated, and the ends to 
be sought, determine the rational procedures available to any society; 
( d ) it is legitimate, in the absence of direct empirical evidence to the 
contrary, to impute to societies the rational procedures available to 
them by virtue of their known (or postulated) sociocultural levels. 

146 • Chandler Morse 

is ignored or assumed away. If, for example, one is prepared to 
accept the premise of neo-classical economics that all economic 
behavior represents the pursuit of individual goals, where each 
actor attempts to maximize his personal values and takes rational 
account of the consequences of action only for himself;* and if one 
therefore regards the economic system as governed by "consumers' 
sovereignty," the embodiment of common social and cultural ele- 
ments in personality systems suffices to establish the presumption 
that a certain orderliness can be achieved in the operation of 
society without collective action in pursuit of social goals. But in 
this case, goal-attainment and adaptation would be functional 
imperatives of personality systems only, not of social systems. The 
only sense in which one could speak of the attainment of social 
goals, of maximizing social welfare, would be in the sense of some 
aggregate of individual goals. This is precisely the sense in which 
these terms are used in modern welfare economics; and the prob- 
lem of definition and aggregation has been a fruitful source of 
difficulty and controversy.! 

Parsons opposes the foregoing view unequivocally. The Structure 
of Social Action is an extended essay designed to show that neither 
an individualistic-hedonistic nor a mechanistic theory of action will 
do. Econoraij and Society is more explicit: 

The goal of the economy is not simply the production of income 
for the utility of an aggregate of individuals. It is the maximiza- 
tion of production relative to the whole complex of institutional- 
ized value-systems and functions of the society and its sub- 
systems. As a matter of fact, if we view the goal of the economy 
as defined strictly by socially structured goals, it becomes inap- 
propriate even to refer to utility at this level in terms of indi- 
vidual preference lists. . . . The categories of wealth, utility, and 
income are states or properties of social systems and their units 
and do not apply to the personality of the individual except 
through the social system. (ES 22; aU italics added except the 

* To "take rational account of consequences" is to avoid costs and 
seek benefits for the appropriate reference system, whether an in- 
dividual or group. The neo-classical economic premise excludes the 
possibility that a group might be an appropriate reference system, 
f See, for example, Tibor Scitovsky, "The State of Welfare Eco- 
nomics," American Economic Review, Vol. XLI (1951), pp. 302-15. 


The proposition that social systems can have goals, and can 
seek facilities for their attainment (adaptation), is implicit in the 
fourfold schema and very nearly explicit in the foregoing passage. 
What is not explicit is the transition from individual ends and 
means to social ends and means; from a concept and explanation 
of individual behavior to a concept and explanation of collective 
behavior. The opposition between the individualistic and the col- 
lectivistic views of social acton is posed but the reconciliation is 
incomplete. It is incomplete because no attempt has been made 
to put meaning into the concepts of social ends and social means. 

To put it briefly, goal-seeking takes on social character when 
and to the extent that the actor ( individual or group ) takes rational 
account of the consequences of his action for others than himself. 
A. C. Pigou made use of this notion in his distinction between 
"private" and "social" methods of calculating economic benefits 
(ends) and costs (means), but the principle, though frequently 
employed in economic analysis, appears to have far wider implica- 
tions and applications than have yet been developed.* Thus, 
Parsons' "self-collectivity dilemma," though posed as a problem of 
deciding whether to treat other actors as members or as non- 
members of a given social system, can readily be broadened to 
cover the problem of whether to take wide or limited account of 
consequences— that is, whether to pursue a more individualistic 
or a more collectivistic course of action. The individual then stands 
at one end of a "self-collectivity" continuum with all mankind at 
the other end.f 

In organized societies, where the number of decisions that any 
individual can make without aflFecting others is exceedingly limited, 
the self-collectivity dilemma is pervasive. Societies that are organ- 
ized on a primitive, communalistic basis resolve the dilemma to 
a large extent by "pre-deciding" it and institutionalizing the deci- 
sion in various ways. More complex societies have not solved the 

* A. C. Pigou, Economics of Welfare, 4th ed. (London: Macmillan, 
1932). For examples of empirical applications of the distinction be- 
tween private and social benefits and costs see K. W. Kapp, The Social 
Costs of Private Enterprise (Cambridge, Mass.: Harvard University 
Press, 1950). 
f Compare Tom Paine: "My country is the world." 

148 • Chandler Morse 

problem in this way— perhaps they cannot— and the fact that they 
have not raises certain fundamental issues which bulk large in 
Marxian analysis but are ignored or slighted by Parsons. 

These issues concern the need for social systems to embody: 
(1) an ideology that defines eflFective social ends and appropriate 
social means; (2) a set of hierarchically superior roles, with respon- 
sibility, authority, and power to make the ideology effective; (3) 
criteria for determining the eligibility of candidates for these roles; 
(4) standards by which to select particular role incumbents from 
among the eligible candidates; (5) standards by which to allocate 
rewards among incumbents of top social roles vis-a-vis other roles; 
(6) mechanisms by which to apply the ideology, that is, to coordi- 
nate behavior throughout the social system (or sub-system) in the 
interest of employing the appropriate means for the attainment of 
the effective social goals. Corollary issues involve the extent to which 
different ideologies, social goals, and hierarchies can coexist within 
the same social system; the extent of the need for compatibility 
between the subordinate ideologies, goals, and hierarchies of sub- 
systems with those of superordinate systems and with each other; 
and the processes by which such ideologies, goals, and hierarchies 

Parsons perhaps considers that he has dealt with these issues. 
What he calls the dominant value system of a society could be held 
to constitute its ruling ideology, and the hierarchy of values by 
which performances and qualities are ranked could be said to 
establish the hierarchy of roles that forms the basis of social class 
distinctions.* Consider also his statement that the 

. . . business community is responsible for a conspicuous output 
of ideological matter which expresses business leaders' concern 
for matters of organizational responsibility with special reference 
to the "principles" on which the whole economy is organized. 
This concern we believe derives from the integrative and value 
aspects of their roles, not primarily from their "economic inter- 
ests" in the usual sense. (ES 151) 

* See his article, "A Revised Analytical Approach to the Theory of 
Social Stratification," in Bendix and Lipset, eds., Class, Status, and 
Power (Glencoe, 111.: Free Press, 1953), pp. 92-128. 


Furthermore, allocation to the G-sub-system of the function of 
mobilizing power, and its designation as "Polity" at the societal 
level, suggest that Parsons regarded this sub-system as standing 
in a hierarchical relation to the rest. So does the statement that 
"the focus of the executive role as v^e ordinarily understand it, is 
on responsibility for system goal attainment."* But there is a strong 
implication in the Administrative Science Quarterly articles, and 
even more in Smelser's new study,j that executive decision-making 
roles are to be found in each of the sub-systems (except, possibly, 
Latency ) . 

Most of this is extremely sketchy. Moreover, there are several 

First, the existence of social goals and of social ( collective ) goal- 
seeking behavior is taken for granted, with no specification of 
how social goals get defined, or of how social goal-seeking differs 
from individual. It is implicitly assumed that the means-ends schema 
can be taken over from individual behavior and applied to social 
behavior without justification or explanation. In Working Papers 
goal attainment means "consummatory gratification," a concept of 
individual psychology that strikes one as only mildly inappropriate 
when carried over to the analysis of small group behavior. But in 
Economy and Society, goal attainment has become the "capacity 
of the society to attain its system goals, i.e., collective goals," where 
capacity is "power as distinguished from wealth." (ES 48) Indi- 
vidual psychology has been pushed aside but no clue is provided 
concerning the steps by which transition to the new conceptualiza- 
tion was effected. 

Second, the possibility that a "social class" might also be a con- 
crete social system, with functional needs and sub-systems of its 
own, receives no consideration. Such a class (which need not be 
hereditary) may be dominated by individuals who occupy roles 
with the same functional primacy, but it would not be identical 
to any of the functional sub-systems (G, A, I, or L). 

Third, there is a tendency to widen the spread of decision-making 
roles, from an initial concentration in the G-sub-system to a distri* 

* Ibid., p. 114. 

f Smelser, Social Change in the Industrial Revolution, pp. 24flF. 

150 • Chandler Morse 

bution over three and perhaps all four of the sub-systems. This 
appears to have sprung from a recognition that social goal-seeking 
requires superior coordination to supplement the built-in types of 
coordination described in Section II. At times there has been a 
tendency to identify such coordination with the integrative function 
( ES 48-9 ) ; at other times to suggest that the integrative problem is 
concerned only with coordination in the interests of system-mainte- 
nance, leaving that required in the interests of task-performance to 
be provided otherwise (by the G-sub-system? ) . 

A fourth difficulty, correlative with the third, is the problem 
of locating technical role behavior, particularly that associated 
with worker roles. This problem was ignored up to and including 
Economy and Society, but in Smelser * these roles are located in 
the Latency sub-system of the industry ( and Economy, by implica- 
tion). No hint is given of how to square this with the concept of 
Latency as developed in Working Papers and even in Economy 
and Society. 

Are these and other omissions and ambiguities merely the care- 
less consequence of rapid work? No doubt to a degree they are. But 
they also appear to be the consequence of Parsons' basic concern 
with the sources of order in social systems. A society in which there 
was universal membership in a single collectivity, universal partici- 
pation in a single task-performance process, perfect conformity to 
role expectations, and perfect consequent attainment of all social 
and personal goals might be regarded as the prototype of "perfect 
social order." System-maintenance problems would not arise. Such 
a society would have no need for power over men, but onlv over 
things; no need for superior coordination, but only for built-in 
coordination; no need for a superior ideology or a hierarchy to 
maintain it; and no need for "self-orientation vs. collectivity-orien- 
tation" to define the relations between systems placed in a hier- 
archical order. Parsons clearly does not intend to imply that actual 

* Op. cit., pp. 24 and 45. The fact that such roles are mainly con- 
cerned with transformation processes rather than transactions may 
have been justification for ignoring them in earHer work. There is 
much to be said for regarding technical role behavior as simply the 
means by which role responsibihties are discharged, and therefore as 
external to the social system. 


societies conform to these specifications of perfect social order. He 
believes that maintenance of order is a major problem, but his 
belief that order cannot be maintained solely by coercion leads him 
to stress the importance of the internalized ordering processes of 
system-maintenance. This stress, which might be regarded as the 
antithesis of Marx's thesis of social conflict (in noncommunistic 
societies), leads Parsons to regard power as an interesting side- 
phenomenon rather than a central feature of social systems. 

It is probably more correct to regard power as a major element 
in its own right, possibly tied to a fifth functional imperative- 
superior coordination.* Had he regarded power in this way. Par- 
sons might have seen that effective social goals are of necessity 
those that are desirable from the standpoint of the powerful. This 
might have led him to see that hierarchical definition and enforce- 
ment of social goals diminishes in proportion as the distribution of 
power is equalized, and to inquire into the process by which the 
devolution of power from the few to the many changes the char- 
acter of effective social goals, bringing them more into line with 
an equalitarian concept of the public interest and the general 
welfare, and moving society an important step closer to "per- 
fection" of social order. He might then have seen in this a clue 
to the meaning of policy-determination, as something that must 
stand outside and above the essentially administrative processes of 
task-performance and system-maintenance. A hierarchy of com- 
mand to coordinate complexly differentiated processes, to assure 
consistency of intermediate goals with each other and especially 
with ultimate goals, is a technical necessity: It requires superiority 
of position (role relationship) and of authority, this being the 
essence of an administrative hierarchy. But it is an open question 
whether or to what extent the selection or legitimation of the goals 
to be sought by an administrative hierarchy requires the existence 
of a social hierarchy, including a power structure that enables the 

* There is hint of this possibility in the reference to the four im- 
peratives "plus the factor of relative importance or 'weight' [= power; 
FSI, 45, 75, 151] of a unit in a system," as the basic variables of a 
system. (ES 37) Consider also the reference to the "superiority and 
power" of the position of the executive within an organization. (ES 

152 • Chandler Morse 

higher classes to determine social goals governing all classes. That 
administrative and social hierarchies tend to be found in mutually 
dependent association is an empirical fact that may reflect imper- 
fection of social integration rather than ineluctable necessity. In 
any case, it is clear that the methods of policy-determination, the 
character of political activity, and the organization of the polity 
vary according to the kind and degree of social stratification. 

The foregoing distinction between administration and pohcy— 
that is, between task-performance and system-maintenance, on the 
one hand, and over-all direction of society as a whole on the other 
—would require some revision of Parsonian theory but would seem 
to introduce no insoluble problems. On the contrary, it appears 
likely to resolve certain difficulties and, perhaps, to point toward 
a more effective theoretical synthesis of the interplay of cooperation 
and conflict in human affairs. 



Alfred L. Baldwin 







The theory of personality developed by Parsons is 
not merely another variety of the perennial crop of per- 
sonality theories. It has a genuine new look and explores a 
dimension not commonly found in psychological theories. 
Its novelty as well as some of its preoccupations seems to 
reflect Parsons' approach to psychological problems from 
the functional viewpoint in sociology. 

One consequence of his approach is that personality 
theory is not intrinsically very important to him. His real 
commitment is to the problem of stabiHty and change in 
a complex social system, not to the conceptualization of 
individual personality. Nobody, however, who is attempt- 
ing an inclusive theory of the whole of social science can 

154 • Alfred L. Baldwin 

ignore the personality of the individual, because the social system 
operates and functions through the behavior of individuals. The 
individual is the cog— or the monkey wrench— in the social ma- 

From such a viewpoint the most relevant features of the indi- 
vidual personality are those that affect his social functioning. If 
there are psychological differences that make no difference as 
far as the social system is concerned, their investigation is quite 
properly to be left to those with more individualistic interests. 
The psychologist looking at Parsons may therefore find some of 
his own favorite distinctions and controversies ignored; or he 
may find them handled in cavalier fashion. On the whole, how- 
ever, the psychologist should refrain from showing his teeth over 
such issues. Instead, he should direct his attention to those aspects 
of the theory that are most relevant to the conceptualization of 
the individual in the social system. 

A second consequence of being a sociologist is that the funda- 
mental problem of personality theory looks different from the way 
it does to the personality psychologist. Anybody who looks at 
society in any detail must have the impression that the coordina- 
tion of individual efforts into a smoothly functioning society is 
a stupendous task, and the machinery of coordination seems 
remarkably loose, fumbling, and vulnerable to individual whim. 
One is tempted to conclude, "It can't work— it can't possibly work." 

It is clear that a major job of coordination is involved in main- 
taining the social order. No group of individuals acting at random 
could supply themselves with food, clothing and shelter, let alone 
providing for procreation of the society and its maintenance in 
the face of hostility. Even if all the individuals were merely cogs, 
the task would be immense. 

But the functional sociologists insist— and the facts support them 
—that the task is made even more difficult by the unfortunate 
recalcitrance of the individual personality. Maintaining the social 
system is somewhat like keeping an unruly mob in some semblance 
of order. It is achieved by a variety of devices: providing individ- 
ual rewards for collectivity-oriented behavior; imposing punish- 
ments for social deviance; establishing a social-emotional specialist 
whose job it is to maintain individual cooperativeness and morale 


when it is strained by the necessities of group task activity; allow- 
ing holidays of individualistic goal gratification following periods 
of task-oriented behavior; and instituting phases of inspirational 
retreat wherein individual commitments to group values are tem- 
porarily heightened and polished before another sally into the 
cold cruel world. 

Anybody who began the study of personality after gazing 
intently at such a spectacle would find his point of view colored 
by the experience. He would wonder whence comes this intran- 
sigence of the individual actor; is it original sin, short-sighted 
stupidity, or bad preparation during childhood? Perhaps even 
more important, why doesn't the whole system blow up in our 
faces? The fact that these are not the questions initially raised 
by the usual personality theorist makes a big difference in the 
kind of theory that emerges. 

If his functionalism is a strong factor shaping the theor}^ of 
Parsons, there is also a more idiosyncratic factor that plays an 
equally important role. Besides being a sociologist. Parsons might 
be called a unitary isomorphist. He sees all the phenomena of the 
social world, institutions, societies, personalities, processes, defenses 
and taxonomic classifications as formally isomorphic to each other. 
In this same pattern, he pictures almost every phenomenon as a 
nest of boxes containing parts within parts within parts, each 
isomorphic to the whole. Such a conviction appears on many occa- 
sions to bring Parsons into conflict with facts, but he sticks to 
his guns and is never satisfied to leave a personality concept with- 
out its parallel in the social system. 


If we accept conflict and reconciliation between individ- 
ual and society as a basic problem for personality theory, it would 
be well to consider how such a problem might be theoretically 
described. A survey of alternative forms of theoretical representa- 
tion will make it clear how Parsons sees the problem, although this 
survey is not itself a part of Parsonian theory. 

156 • Alfred L. Baldwin 

In theories of behavior, conflict of the person with society is 
usually represented as a conflict with the environment— we will 
come later to the special problems of representing the environment 
as a social one. To make the problem extremely simple at first, 
consider the behavior of a blind bumbling organism whose actions 
are restrained only by actual physical fences, or other absolute 
barriers. Furthermore, endow this organism with a single simple 
motivational mechanism, a set of desires for various types of goal 
objects found in the environment. Notice that already restrictions 
have been located in the external world and motivation in the 

Actually, this situation could be represented in three ways. 

1. The individual could be endowed with desires and abilities 
and his behavior could be seen as his attempts to gratify his desires 
in the face of his own weaknesses. In other words, the conflict 
could be made internal to the person. In this case, the environment 
would be described in terms of the properties that made certain 
actions difiicult or easy, desirable or repulsive. 

2. The environment might be pictured as containing barriers and 
impositions on the one hand, and valences (attractive or unat- 
tractive) on the other. Now the person is conceived as merely a 
point of reference and the conflict lies in the fact that some of the 
positive valences are inaccessible. Both the motivations and the 
restrictions are attributed to the environment. 

3. The environment might be represented as containing the bar- 
riers, and the person represented as having the needs. Now the 
person is pictured as struggling against the environment to satisfy 
his needs. Probably this model resembles most closely the "man in 
the street's" view of the world. We generally tend to see ourselves 
as motivated and the environment as offering helps or hindrances. 
This picture will be especially suitable if the barriers of the 
environment are so strong as to be impassable to anyone, i.e., essen- 
tially the same for all people, while different people strive for dif- 
ferent goals, so that the valences differ from one person to another. 

The same three basic alternatives still exist when new sorts of 
causal factors are introduced. If some of the restrictions upon 
behavior or some of the impositions upon the individual operate 
through sanctions rather than through actual barriers, then certain 


markers or signs in the environment act as if they were actual 
restraints upon behavior. For the behavior theorist, these restraints 
may be located in the environment, and not in the person; or they 
may be conceptualized as needs of the person to avoid punishment 
or attain reward. In the latter case, the struggle of the individual 
against the restraints of society is viewed as a conflict of motives. 
Again naive behavior theory tends to attribute the restraints on be- 
havior to the environment, especially if they are strong enough to 
restrain everybody. In everyday language, the individual may be 
forced to behave through fear for his life and not be considered re- 
sponsible for his acts; he is not, however, relieved of his responsibility 
by the existence of a very large reward contingent upon his behavior. 
We do not assume that every man has his price. Moral imperatives 
and taboos are especially likely to be attributed to the environment 
because they are shared by almost all of the society and are morally 
absolute, i.e., are to be respected regardless of the strength of the 
opposing motivation. 

No single one of these views of the person in the environment is 
"right." In fact, if one is rigorous in his conceptualizing they are 
all three wrong. But each remains useful for certain purposes. The 
danger lies in the fruitless controversy that may stem from the 
tacit acceptance of diflFerent models by different people. Thus, an 
argument that sometimes arises between sociologists and psycholo- 
gists is whether "role" is a personality concept or an environmental 
concept. For many purposes, it is convenient to attribute it to the 
environment since it is a stable feature of behavior that elicits the 
same behavior from all who have that role. For other purposes it 
is convenient to think of role as a personality characteristic. 

It is important, however, not to mix the two viewpoints haphaz- 
ardly. A serious diflBculty a personality psychologist finds in Parsons' 
writings is the vagueness with which Parsons treats these theo- 
retical issues. While his intent seems reasonably clear, namely, to 
view the personality as an interdependent set of need-dispositions, 
internalized social objects, role-expectations and values, he unfor- 
tunately says something quite different whenever he is explicit 
on the matter. This unfortunate conflict between Parsons' intention 
and performance will become increasingly apparent as the theo- 
retical system is presented. 

158 • Alfred L. Baldwin 


The concept which must turnish the starting point for an 
exphcation of the Parsonian theory of personaHty is need-disposi- 
tion. Need-dispositions are the fundamental units of the person- 
aHty system. In Parsons' formal description of his theoretical 
concepts (GTA), need-dispositions are in fact the only units in 
the system. In the isomorphism between the personality system 
and the social system, need-dispositions correspond to individual 
people. Thus the essential conceptual model seems to be of the 
personality as a set of individual need-dispositions whose gratifica- 
tions are neither entirely compatible with each other nor wholly 
possible within the impositions of the environment. These units 
are integrated, coordinated, and modified by value standards, role- 
expectations and the like, in the interests of maintaining the 
personality system and optimizing gratification within the limita- 
tions of the environment. 

What is a need-disposition? The hyphenization is intended to 
suggest that it involves both an activity (a performance) and a 
type of satisfaction (a sanction). Hunger, for example, involves 
eating as an activity and the gratification that comes with it. 
Parsons does not go into more detail; the implication seems to 
be that the performance aspect of a need-disposition is the con- 
summatory activity associated with gratification. Sometimes, how- 
ever, Parsons suggests that the performance is an instrumental 
action for obtaining the satisfaction, as for example when achieve- 
ment is the performance and approval the sanction, but presum- 
ably unless the achievement itself is gratifying in the same sense 
that eating is gratifying, we would not label achievement a 

A need-disposition reflects a categorization both of actions and 
of environmental events. All members of one class of acts are 
called examples of the same performance, and all members of an 
associated class of environmental events are called the same 
sanction. Personality theorists have come to no consensus on the 
appropriate methods of categorizing these events. The most com- 
mon motivational unit is the need, which seems essentially like 


need-disposition as Parsons uses it. The basis for the categorization 
o£ the acts and events is more intuitive than expHcit; usually the 
basis of categorization is some vague similarity of the act and the 
associated events. Dependence, for example, is marked by ego 
being influenced by alter, asking alter's advice, etc., all of which 
are viewed as having some inherent similarity. In the usual set of 
needs, the need does not specify the alter. Dependence as a need 
for dependent relationships usually connotes that it is the nature of 
the relationship rather than the particular alter that is essential 
for its satisfaction. 

If we look at the types of need-dispositions Parsons describes, 
we shall see some of the theoretical problems that are involved. 
One type of need-disposition relating to social objects is exemplified 
by the need for esteem, approval, response, or love. Consider the 
activities that reflect ego's love for alter. Ego wants contact with 
alter, both physically and through intellectual exchange, and he 
wants emotional understanding. If alter wants some goal object, 
ego's love for alter makes ego want alter to have that goal. If 
some outside person is hostile to alter, ego is hostile to the out- 
sider. The activities in this range have no particular behavioral 
similarity as do the various acts of eating, or striving for excellence, 
and the like. Instead, the invariant in the set of actions lies in the 
object rather than in the acts themselves. In many theories ego's 
love for alter would be called a sentiment and would be taken as 
sufiicient motivation for the various actions involved. Parsons in- 
stead speaks of a need-disposition for love that is gratified by 
loving alter and being loved by alter. Just what implications, if 
any. Parsons intends by this terminology are not clear. 

In ordinary language, a "need for love" is quite different from 
the "sentiment" of love. A need for love is free-floating; it can 
be satisfied by any love object but does not imply that any actual 
love relationship exists. Colloquially, the need for love is sometimes 
described as being "in love with love." The "sentiment," on the 
other hand, implies an alter and a love relation to alter. A need 
for love is selfish and self-oriented. The sentiment of love implies 
a certain altruism and tender feelings for another person. It is not 
clear how Parsons intends the concepts of love, esteem, and so on, 
to be used. Sometimes he uses them one way, sometimes another. 

160 • Alfred L. Baldwin 

Another type of need-disposition is a value. Here, again, the in- 
variant in the set of performances and sanctions is not easily put 
into words. The value, fairness, for example does not motivate 
a set of actions that have a common result as do eating, striving for 
success, or nurture. Neither is the object the invariant as in the 
case of sentiments. Instead, the invariance lies in the adherence to 
a principle or rule of behavior and reflects the presence in the 
actor of the acceptance of certain "oughts." Parsons seems some- 
times to mean by "value" a "need for conformity to values stand- 
ards" but unfortunately he is not very clear on the point. At 
another point, he describes a value as a need for end states 
that are demanded by a value standard. In other words, we cannot 
be sure whether Parsons means a need to conform or needs that 
are, in fact, compatible with value standards, or both. 

A third type of need-disposition comprises role-expectations. 
These are "needs to get 'proper' responses from alter and disposi- 
tions to give proper responses to alter." It appears later, however, 
in Family, Socialization and Interaction Process (1955), that a 
role is not a need-disposition as such but is a sub-system of the 
personality, motivated by a number of needs and providing the 
necessary integration or fusion of the component need-dispositions. 

Already, we can see what is a basic conceptual unclearness in 
Parsons' formulations. He cannot really decide whether or not 
he wants to picture the personality as a set of need-dispositions 
whose aims are segmental (i.e., self-oriented) but which are inte- 
grated within the personality by sets of values and role-expectations. 
There are many attractive features in such a model. It emphasizes 
the parallelism between the personality system and the social 
system. It corresponds to naive behavior theory that pictures ob- 
ligations as controlling wants. It is very close to psychoanalytic 
theory, with many of whose concepts Parsons agrees. 

On the other hand, such a formulation has real difficulties. It 
seems to deny the basic need-disposition character to such impor- 
tant motivations as love, nurturance, group loyalty, and so on. 
These are not segmental but essentially integrative. To view them 
merely as controls over more segmental needs but not needs them- 
selves seems to put social integration primarily upon a basis of 
"enlightened self-interest," because all the fundamental gratifica- 


tions in such a model are private and individualistic. Social systems 
depending largely upon this type of control are, in the opinion of 
some sociologists, those characterized by "anomie." Parsons him- 
self, in his article on the superego ( WP ch. 1 ) , goes to considerable 
lengths to point to the importance of socialized elements even in 
the id. 

Perhaps through such considerations as these. Parsons is almost 
forced to give the status of need-dispositions to altruistic senti- 
ments, values, and role-expectations. But in so doing, he blurs the 
neatness of the conceptual model of the person versus environment. 
Furthermore, he is in danger of picturing a social system that does 
not need external controls at all because they are all part of the 
individual's motivation. This suggests a Utopian condition contra- 
dicted by the facts of life. 

Parsons is quite right, of course, in trying to avoid either of 
these black or white alternatives. An adequate theory must some- 
how synthesize the two. Some social integration and control is 
based upon sanctions of segmental need-dispositions; some is medi- 
ated through value commitments resembling the superego working 
through guilt feelings aroused by violation of the standards; some 
is built into the basic need-dispositions of the individual. There 
is no reason why a theory of personality should not include all of 
these mechanisms, but it should distinguish among them. Parsons 
seems to classify them all together. Yet, the differences in the social 
integrative behavior of people whose ties to the collectivity are 
mediated through these different psychological mechanisms will 
make all the difference in the world in the stability of their con- 
tributions to the collectivity (see p. 188). 

Oddly enough, it might be argued that the lack of differentiation 
in Parsons' concepts of the psychological mechanisms underlying 
social controls lies in his failure to consider deviance. He pictures 
deviance in such an institutionalized manner that there is little 
room for genuine individual difference. Parsons describes the child 
who is suddenly confronted with new socialization demands as 
deviant— by fiat so to speak— and his natural reaction to the imposi- 
tion tends to make him more deviant, but these reactions are the 
modal ones and Parsons does not discuss how the individual child's 
response to socialization might depend on whether his love for 

162 • Alfred L. Baldwin 

his mother is contingent upon her gratification of self -oriented needs 
or is firmly rooted in the need-dispositional structure of the child. 


How do the need-dispositions and other aspects of the 
personality develop? Parsons advances three propositions about 
the socialization process. The first is that the socialization process, 
analogous to psychotherapy, group learning, and social control, 
goes through the phases Latency, Integration, Goal-gratification, 
and Adaptation in that order. Figure 6 combines Figures 1 and 2 
in Family, Socialization and Interaction Process. 

This diagram fits the phases of social control into the Freudian 
stages of psychosexual development, but before looking at the proc- 
ess of socialization itself, we should consider how Parsons describes 
the process of social control. The basic reference point is psycho- 
therapy. The process begins in the L cell because that is the phase 
primarily involved in "tension management." The patient is in a 
state of high tension because his values are in conflict with those 
of the collectivity. The first task of the therapist is to be permissive 
of the patient's expressions of his deviance. The therapist does not 
respond with the usual sanctions to expressions of deviant values, 
or to symbolic expressions of aggression, dependency, and other 
inappropriate needs, but instead permits the patient to talk about 
them freely. This creates transference, so that the next phase of 
the process is integrative, i.e., is concerned with the interpersonal 
relationship. During the I-phase the therapist provides support and 
expressions of acceptance. He does not withdraw from the relation- 
ship, either because of the deviance or the dependency of the 
patient upon him. The therapist does, however, deny reciprocity 
( G phase ) . He does not love the patient in response to the patient's 
love. He does not concur with the patient's deviant values. In other 
words, he does not permit actual goal-gratification in the inter- 
personal relationship. He begins to exert pressure to bring the 
patient back into adjustment to the realities of social life rather 
than being seduced by the patient into joining him in deviant 
collectivity. This is all the G phase. Next, the therapist begins to 



reward and punish (A phase) and really puts on the pressure to 
modify deviant behavior— not so much by actual manipulation of 
gratifications as by representing clearly to the patient the realistic 
consequences of his actions. Because of the solidarity built up 


a. Adaptive-Instrumental 

a. Instrumental- Expressive 
( Goal-gratification) 

b. Mampulation of Rewards 

b. Denial of Reciprocity 

c. Maturity (genitality) 

c. Latency (psycho analytic) 
(4 objective system) 


Enter here 

CRISIS: di ' 

a. Latent 

a. Integrative 

b. Permissiveness 

b. Support 

c. Oral Dependency 

(Mother- Child Identity) 

c. Love attachment 
( Parent-Self object 

PHASE: da 


a. Task-performance phases (AGIL) 

b. Learning-social control phases (LIGA) 
0. Phases of psychosexual development 

d. Crises of transition 

Figure 6. Phases of psychosexual development. 

during the permissive and support phases, the therapist can now 
actively modify the patient's behavior and help the patient to come 
to a realistic adaptation to the demands of his situation. 

The accuracy of this description of psychotherapy is not easy 
to assess. Therapists would doubtless feel it was higlily schematic 


164 • Alfred L. Baldwin 

and oversimplified. There are, however, certain features of the 
description that seem accurate, especially the necessity of estab- 
lishing a strong, affectionate and dependent relationship that is not 
threatened by the patient's expressions of immoral, antisocial, or 
hostile wishes. This relationship is a foundation for facing the 
patient with reality and helping him adapt to it. 

It is this feature of psychotherapy that is most clearly analogous 
to current views of socialization. The parent builds up a strong 
love and dependency relationship with the child through early 
indulgence and gratification. This solid foundation of credit permits 
the parent later to withhold rewards and impose punishments with- 
out destroying the child's love for the parent. The conception was 
first developed by Freud and has gained wide acceptance. ^Vhethe^ 
there is a good fit in the details of the analogy between socialization 
and psychotherapy is more questionable. One important difference 
is that during socialization the parent himself puts the child in a 
•deviant position by shifting his demands; then he must rebuild 
the love imperiled by his action without retraction of his demands. 

The usual psychoanalytic picture of infancy begins with oral 
dependency and is intended merely to record the facts that the 
child is at first only physically dependent upon the mother, but 
that by lavish indulgence of liis needs— particularly his sucking 
needs— the infant becomes psychologically dependent. This depend- 
ency is first threatened by the reduction in unlimited gratification 
that occurs late in the first year. This restriction on gratification 
frequently occurs at weaning. 

Parsons' way of describing the same process illustrates his general 
conception of socialization. The neonate is an aggregate of seg- 
mental biological needs, motivationally speaking. The mother first 
cares for these needs; then during the oral crisis, she denies the 
■child's demands for continuous and exclusive gratification of each 
separate need but rewards a more diffuse dependency relationship. 
As a result, the heretofore independent segmental needs are fused 
into a dependency upon the mother in which individual needs 
are subordinate to the collectivity (i.e., the personality system). In 
the usual Parsonian scheme of orbits within orbits, he describes 
this little cycle of socialization as an L-I-G-A sequence within the 
L. phase of psychosexual development.. 


This language of Parsons' points to an aspect of the development 
of dependence that is not obvious in Freud's description. It is an 
interesting viewpoint that clearly stems from Parsons' general 
interests. When examined in detail, however, its relation to the 
general theory is not obvious. For example, it is not obvious why 
such a process should be attributed to the latencv rather than 
integrative phase or why the mother's behavior should be consid- 
ered an example of permissiveness rather than support. 

A second general principle underlying the Parsonian theory of 
personality development is founded upon the assumption that when 
the child establishes a relationship with a social object, the rela- 
tionship itself consists of a social system in which each member 
has a certain role with expected performances and sanctions. The 
common notion among psychologists is that the child learns his 
own role in this system, acquires some attitude toward the other 
person, and perhaps accepts the values of alter. But Parsons sug- 
gests that the child internalizes the whole system, so to speak, and 
acquires the performances, need-dispositions, and attitudes of all 
of the members of the system. 

The second phase of socialization illustrates this principle. It 
is another cycle of socialization beginning where the first left ofiF, 
with oral dependency. In it the mother requires autonomy from the 
child and rewards it with love. The result is the Love-Dependency 
Personality or the Two-Object-System in Figure 7, taken from 
Figure 7 in Family, Socialization and Interaction Process. The 
mother-child social system is marked by a diflFerentiation primarily 
•on the power dimension, but also on the instrumental-expressive 
dimension. The two need-dispositions that emerge represent in one 
case the role of the mother toward the child, and in the other the 
Tole of the child vis-a-vis the mother. As the child internalizes 
alter, he also internalizes alter's orientation toward himself as an 
object, subject to the possibility of the child's not perceiving this 
orientation entirely correctly. In summary: 

The internalized personality establishment, therefore, though 
originally built up through the experiencing of functions per- 
formed for ego by an alter can from then on always serve as an 
agency of the performance of the same functions either in return 
for alter or in the role of alter for ego himself. (FSI 74) 


Alfred L. Baldwin 



Cathected: Self 
Internalized: Parent 

Need-disposition: Dependency 

Performance Type: 

Alter-oriented: Asking for and giving care 
Narcissistic: Self-indulgence 

Sanction Type: 

Alter-oriented: Accepting care 
Narcissistic: Self- gratification 



Cathected: Parent 
Internalized: Self 

Need-disposition: Autonomy 

Performance T3T5e: 

Alter-oriented: Loving alter 
Narcissistic: Self-love 

Sanction Type: 

Alter-oriented: Receiving alter's love 
Narcissistic: Self-love 

Figure 7. Second phase of personality structure: love-dependence 

This is an interesting and suggestive way of describing the anal 
stage, but if we look at the example of the two-object system 
illustrated in Figure 7, we find a really hopeless hodgepodge of 
theoretical inconsistencies. First, we see the need-disposition that 
emerges from internalizing the mother's role is not nurturance but 


dependency. The performance correlated with this need is both 
asking for care and giving care. These are surely not the same 
need-dispositions. The sanction is "accepting care," which does not 
sound like a sanction at all, unless it is alter's accepting ego's care, 
and in those cases the need-disposition is autonomy while the 
role performance is loving alter and the sanction is receiving love. 
How can the performance of "autonomy" be "loving alter"? Parsons 
is apparently led to such a position by the argument that the 
child's autonomy is rewarded by the mother's love. But an instru- 
mental act cannot be used to label the consummatory behavior 
without creating all kinds of confusion. Another signficant clue to 
why autonomy and love are mixed is that in the next stage auton- 
omy and love are going to be differentiated, the two new need- 
dispositions being adequacy and security. 

Moving on to the next phase, Parsons argues that the two-object 
system of mother and child differentiates into the four-object system 
consisting of father, mother, brother, and sister by means of a 
differentiation of the power from the instrumental-expressive 
dimension. The post-oedipal personality structure takes the form 
shown in Figure 8. Several curious features also appear in this 
stage. The need-disposition conformity has as its performance, 
control of alter; and the sanction is esteem. Despite this definition, 
conformity is equated to superego. 

All through this portion of the argument, it is almost impossible 
to escape the feeling that Parsons makes the developmental process 
fit the theoretical model only by straining the normal meaning of 
terms beyond reason. Even where there is no flagrant violation of 
usual meanings, there are frequent shifts in connotation. For 
example, let us trace the history of the terms instrumental, expres- 
sive-adaptive, and integrative, together with the associated pattern 
variables, from the Working Papers through to Family, Socializa- 
tion and Interaction Process. 

While expressive was first used by Bales to describe the kind of 
behavior categorized as "showing tension," this meaning was ex- 
plicitly changed in a later working paper to the usage of the terms 
given in Figure 9. 

In Figure 9 the four terms are used quite consistently. Instru- 
-mental, characterized by specificity and performance criteria, seems 



Alfred L. Baldwin 







Cathected: Self (naasculine) 

Cathected: Self (feminine) 

fiiternalized: Father 

Internalized: Mother 

Need-disposition: Conformity 

Need-disposition: Nurturance 

External Orientation: 

External Orientation: 

Performance: Control of 

P-Giving pleasure 



Sanction: Esteem 

Internal Orientation: 

toternal Orientation: 


P- Self- indulgence 

S- Self-esteem 

S- Se If-gr atif ic ation 



Cathected: Father) 

Cathected: Mother 

Internalized: Self(M) 

Internalized: Self (F) 

Need- disposition: Adequacy 

Need-disposition: Security 

External Orientation: 

External Orientation: 

P- Instrumental Perform- 

P-Giving love 



S- Approval 

Internal Orientation: 

liternal Orientation: 

P-"Reality testing" 

P-Harmoniz ation 

S- Self- approval 


Adaptive Functions Ego Integrative Functions 

Figure 8. The post-Oedipal personality structure. 

to carry the connotation of impersonal. Objects may be dealt with 
impersonally either in adaptive behavior or in consummatory grati- 
fication. Thus, instrumental describes both the A and G cells. 
Integrative is the opposite of instrumental and seems to mean 
dealing with people. It is marked by difftiseness and qualitij orien- 
tation. Qualities need not be limited to ascribed traits, as Parsons 





Adaptive— lastrnmental: 

iistrumental- Expressive: 

Object Manipulation 

Consunmiatory Per- 

formance and Gratifi- 

Perf: Achievement 



Perf: Appreciation 

Sane: Approval Values 



Sane: Response Values 




Latent- Receptive: 

Integratlve-Expresslve : 

Meaning Integration and 

Sign Manipulation 

energy regulation 

Perf: Ascriptive 

Perf: Moral-integrative 



Sane: Esteem Values 

Sane: Acceptance 





Figure 9. 

is inclined to do, but our relations with people depend upon their 
quahties rather than upon their immediate behavior. Cross-cutting 
this dichotomy is the dichotomy adaptive-expressive. In this sense 
adaptive means preparation for an end state, while expressive 
means acting out some internal state. The pairs of pattern variables 
associated with each seem quite reasonable. 


170 • Alfred L. Baldwin 

In Bales' empirical research, he is really concerned with only 
two roles in the small group, the ixistrumental-adaptive on the one 
hand, and the integrative-expressive on the other. The one role is 
concerned with reaching the goal set for the group, the other 
with preventing social disintegration. Both seem adaptive in terms 
of ordinary language usage, one instrumentally and the other 
integratively. The fact that the integrative role was carried out by 
expressing feelings seems little more relevant than that the instru- 
mental role was carried out by expressing opinions. Therefore, the 
tying of adaptive to the one and expressive to the other seems a 
little strained. 

In Family, Socialization and Interaction Process, instrumental- 
adaptive becomes shortened to instrumental and integrative-expres- 
sive becomes shortened to expressive, and they are no longer op- 
posites on the dichotomies originally introduced. To make life 
more complicated, the power dimension is introduced. It obviously 
is not expected to correspond with any of the previous dimensions. 
Yet when the two-by-two table is drawn up, divided on power and 
the instrumental-expressive dimension, the four sanctions— esteem, 
approval, acceptance, and response— are the same four sanctions 
found in Figure 9 when diflFerent dimensions were employed. But, 
look where they appear. Esteem is especially strange; it now is a 
sanction for an instrumental kind of behavior, labeled conformity, 
but whose performance is controlling alter. Previously it was in 
the L sector, adaptive and integrative, and was the sanction for 

This little exercise in textual criticism has been included prima- 
rily to indicate how meanings and interpretations shift and blur 
under what appears to be Parsons' compulsion to make the uni- 
versal pattern repeat itself in every possible circumstance. 

What seems especially unfortunate is that the isomorphism is 
not necessary for the fruitfulness of the conceptualization. The 
requirement that the four sectors of society must correspond to 
the father, mother, brother, and sister in the stylized nuclear family 
seems almost completely gratuitous. In fact, the striking diflFerence 
on the power dimension within the family might suggest that the 
nuclear family is a particularly poor social group to represent 


society. That a semi-permanent group like the family should exhibit 
the same four sectors of activity as a larger society does not seem 
unreasonable, but that each sector should be neatly represented by 
one person's role seems too much to expect. 

In another respect also, the demand for isomorphism seems un- 
warranted. As far as need-dispositions are concerned, there is no 
reason why the child can acquire only four types of need-disposi- 
tions during the first five years of life, even if the whole mechanism 
of acquisition is internalization of social objects in a social system. 
The mother's role in the family encompasses more than one type 
of performance and involves more than one type of sanction. 
Furthermore, she sanctions more than one type of behavior. Thus, 
it does not seem unreasonable that the child might learn both 
love and autonomy during the pre-oedipal stages as two diflFerent 

If a genuine correspondence should emerge between the stages 
of development and the sectors of society, it would certainly be 
interesting and would unquestionably raise a problem for social 
science to explain. But even such a correspondence would be no 
striking demonstration of the validity of the theory whose termi- 
nology made the correspondence apparent. The theory is not tight 
enough to permit truly logical derivations. 

All in all, the game is not worth the candle. In fact, the strenuous 
attempts to realize complete analogy hides the contributions to 
the picture of socialization that Parsons does make. He presents 
many interesting ideas, some of them testable. Some general ones 
have already been mentiond, but the specific ones are also worth 
taking seriously. For example, if there is a detectable distinction 
between the dependency of the oral phase infant and the love 
of the pre-oedipal child that can be attributed to learning au- 
tonomy, it would be interesting, indeed. If, in addition, this could 
be related to the different kinds of identification involved (lack 
of differentiation of mother-child roles in one case, internalization 
of the mother and child roles in the other) it would be quite val- 
uable. But clearly the hypothesis cannot be tested if autonomy is 
to be defined as loving and if dependency means either giving or 
asking for care. There are real values in these formulations, but 
they are in the form of intuitive suggestions, perhaps brilliant ones. 

172 • Alfred L. Baldwin 

The false orderliness and isomorphism is an all too efiFective camou- 
flage for what values are inherent in the general scheme. 

As further illustration of the interesting problems raised by 
Parsons' system at its best, let us turn to a detailed description of 
one cycle of the socialization process employing many of the 
theoretical terms already introduced. 

The cycle is concerned with the transition from the two-object 
system to the four-object system, i.e., the oedipal crisis. The family 
consists of father, mother, the pre-oedipal ego, and an older post- 
oedipal sibling. The different phases of the transition are numbered 
Ti, T2 and so forth. Within each phase each of the different social 
systems within the family is described. 

Ti is the initial stable state. The mother has several roles : family 
member, wife, mother of sib, and mother of ego. 

T2 is the phase in which this stable state is disturbed. Partly it 
occurs because of ego's ability to do more and perform better than 
he has in the past, partly it comes from the family's realization 
that ego is "getting to be a big boy." The shift can be described as 
follows, for each of the social systems involved. 

a. For the family, this total transition is a continuing sequence 
of development to be described as a clockwise cycle from Latency, 
to Adaptive instrumental behavior, to Goal-gratification, to Integra- 
tion, back to Latency. (See Figure 6, p. 163, task performance 
phases.) The first phase, Ti to T2, is the shift from a state of 
latency to the adaptive instrumental phase. The mother as a family 
member shares in this shift of expectations from the child and is 
the agent transmitting the new expectations to the child. In Parsons' 
language she is the communicating link between the family system 
and the mother-child system. 

b. In the mother-child system the disturbance disequilibrates the 
social system and initiates a "therapeutic" cycle of adjustments. 
This cycle includes the same basic phases as the task performance 
cycle, but in reverse order. Parsons labels them permissiveness, 
support, denial of reciprocity, and manipulation of rewards. (See 
Figure 6, p. 163, learning-social control phases.) The disturbance 
in the case at hand is reflected in the mother's withholding part 
of the care that she has customarily been responsible for. Her role 



is thus moved slightly away from the instrumental one as she 
expects the child to take on more of his own care. 

c. In the mother's personality, the cycle goes clockwise, since 
this is a task performance for her, but it may also require some 
adjustments to withhold care from the baby. Because she is anchored 
in the family system, however, and shares its values, and because 
she herself wants to socialize ego, the disturbance is not great. 

d. In the child's personality the cycle goes counterclockwise. 
The disturbance comes as a frustration that deprives him of some 
gratifications, violates his expectations of the mother, and also 
violates his "rights." Within the personality the first reactions to 
this disturbance are: (1) clinging to the original state, (2) wishful 
thinking, (3) removal of the source of disturbance through aggres- 
sion, (4) shifting the internal balance toward narcissistic gratifica- 
tion to balance loss of external gratification, (5) hostility against 
self as a source of disturbance, (6) generalization of expectation of 
disturbance, i.e., anxiety. 

e. In the mother-child system, an integrative crisis impends. The 
mother has created aggression in the child. The child, in response 
to frustration, has become more aggressive and now deviates even 
from previously learned behavior. This is where the permissive 
phase of the psychotherapy paradigm begins. The mother makes 
allowances for the child's deviance; she allows symbolic expressions 
of aggression and dependency wishes but still sticks to her guns 
as far as actual withdrawal of care is concerned. 

f. This now creates in the child a slightly different conception 
of the mother than he had before. She is becoming a different social 
object but still retains her old identity— i.e., it is not as if a strange 
woman replaced the mother in the home. 

T3. This is now the supporting phase of the mother-child rela- 
tions, but it largely corresponds to the latter part of the adaptive- 
instrumental phase in the cycle the family system is following. 

a. In the family system, the mother's behavior toward the child 
is a part of the family policy but it may create a small integrative 
crisis there, requiring the father to provide some encouragement 
and emotional support. The other family members must also back 
her up by making the same requirements of ego that she does. 
Furthermore, the father, as well as the mother, may need to be 

174 • Alfred L. Baldwin 

permissive toward the child if aggression is displaced onto him. In 
Parsons' language this is an input of facilities from the family 
sijstem to the mother subsystem to strengthen her instrumental^ 
adaptive behavior. (A sector). 

b. In the mother's personality, loss o£ love of the child has 
instigated an increase in the expressions of love for the child, 
through frustrations of the mother's need for love and as an instru- 
mental activity to prevent the disintegrative process from getting 
out of hand. Thus, the phase moves to one of support, and also 
we should note that the mother's role has now increased in the 
expressive dimension. Her role is gradually shifting from instru- 
mentality toward expressiveness. 

c. In the child's personality, permissiveness and support have 
partially relieved his anxiety, and probably reduced aggression, 
but have not solved the problem. Information necessary for the 
child to re-define his own role and the social objects has been 
given, but not digested by the child. Parsons is now concerned 
with the process by which the child differentiates the instrumental 
from the expressive role and attaches the instrumental one to the 
father while continuing to see the mother as expressive. He sug- 
gests that in this situation the child cannot directly and immediately 
cathect his "father" as an object distinct from the mother. There 
must be some transition and Parsons argues that the obvious path 
of generalization of the cathexis is by way of some common element 
between the old and the new objects. This necessary element com- 
mon to both the mother and father is the fact that they share the 
"parental" role as well as each filling a special "mother" and 
"father" role. Thus, Parsons seems to say that the child first cathects 
his father as "like mommy" on the basis of the shared responsi- 
bilities of the two, and then gradually understands the father role 
in its unique aspects. 

Next Parsons is concerned with the erotic elements of the child's 
relation to his parents and how this eroticism becomes more 
focused on the mother. "Eroticism" implies a diffuse relationship 
and is not appropriate to describing the relation between the child 
and his caretaker. It emerges, therefore, as the mother gradually 
concentrates on her expressive-integrative functions and turns the 
more instrumental care-taking functions over to the child himself. 


At the same time, these positive feehngs help maintain the 
integration of the mother-child system despite the frustrations im- 
posed by socialization. The same point has been made by Freud 
and others when they point out that a well-established love between 
mother and child makes the socialization efforts of the parent 
more effective. The notion that eroticism emerges as specific care- 
taking decreases is a more novel idea. 

As Parsons points out, 

. . . eroticism is specifically bound to its integrative function in 
the mother-child system. Hence if it is allowed to continue in 
force it will interfere with the integration of the child in the 
family system. This is why every transition to a higher order 
system is marked by a crisis in the erotic sphere, hi each case a 
ladder which has been essential at one stage of the climb must 
be thrown away because it becomes an encumbrance from then 
on. (FSI 210-211) 

T4. a. This is the gratification phase in the task-performance 
cycle for the family system. It is the phase in which ego actually 
treats the father and sibling as objects and expects them to 
reciprocate appropriately. The reciprocation will be primarily re- 
sponse gratification for specific performances in relation to these 
new objects. The father is pleased at the new levels of achievement, 
but selective in the bestowal of rewards. 

b. What from the point of view of the family system is the con- 
summatory phase is for the old mother-child system the phase of 
liquidation. The mother gradually withdraws lower level support 
and substitutes for it "acceptance" on the "parental" level. 

c. In the child the actual fission of the old internalized object 
system takes place. There has been a perception of the mother's 
role in the family (which was only mother and child up to now). 
Now the father is perceived as belonging to the family; secondly, 
he together with the mother constitute a sub-collectivity, parents. 
The father is an object in his own right and in the fission the 
mother-object is not left unchanged. She has become more expres- 
sive and supportive than she was originally, although less so than 
during the transition period. 

T5. a. For the family, this is the integrative phase follo\ving 
gratification, involving the reorganization entailed by the admission 

176 • Alfred L. Baldwin 

of the child to the family system. It involves further acceptance of 
him, and also may require the resolution of rivalries that are 
created by the added recognition given the young child. With the 
completion of this the family returns to the latent phase with 
respect to this particular socialization cycle. 

b. For the old mother-child system T5 is a latency phase in the 
sense that old values are transformed into new ones where neces- 
sary and the residual pre-oedipal attitudes and values are repressed. 

c. For the personality system of the child, it is the adaptive phase 
or consolidation of the new personality structure. 


This detailed discussion makes the picture of socialization 
clear. It is a repetition, over and over again, of the cycle, L-I-G-A, 
from the point of view of the socializing system, and the cycle 
A-G-I-L from the point of view of the child's personality. The two 
cycles go on simultaneously and are cogged into each other so 
that, for example, the goal-gratification phase (G) of the socializ- 
ing system occurs when the socialized system is in the A phase 
(manipulation of rewards). 

Because of correspondence between the phases of the socializing 
and socialized system. Parsons renames the four phases in a 
"socialization" cycle: 1. Primary adaptation, 2. Relative deprivation, 
3. Internalization, 4. Reinforcement. 

From the point of view of psychologists a very interesting feature 
of this theory is the inclusion of the dynamics of the socializing 
system. Parsons asks what inputs are necessary to push this cycle 
around and how it happens that these inputs are available at the 
proper times. Psychologists interested in the individual are likely 
to take the environment as given and to see the cues contained in 
it and the rewards obtainable from it as the ground upon which 
the learning and adjustment process takes place. Parsons, because 
of his sociological viewpoint, sees these cues and rewards as inputs 
which the larger system must provide if the socialization of the 
sub-system is to be completed eflFectively. Furthermore, he asks 
how the socializing system happens to perform its socializing role 


at the right time and in the right phase. The cycle is seen in terms 
of inputs and outputs for each system. 

During the A phase of the cycle, according to Parsons, the super- 
ordinate system provides the sub-system with facilities or aids to 
achievement. This is the input, consisting of information about the 
situation to which the sub-system must adapt. During the G phase, 
the superordinate system provides rewards from the environment 
for correct performance, producing gratification as an output. 
During the I phase, the input consists of narcissistic rewards and 
the output is satisfaction. The terms are obscure, but apparently 
the intent is to describe the kinds of acceptance that build up 
solidarity and enhance personality integration through a diffuse 
rather than segmental sense of satisfaction. Finally, in the L phase, 
the input consists of information again rather than rewards, and 
now information is about the values of the superordinate system 
that tie the sub-system into the larger systems of values. This inte- 
gration into the superordinate system also makes the sub-system's 
own values consistent with each other. 

Naturally, each of these phases involves all four sectors and the 
deprivations incurred in one sector in one phase are balanced out 
in other phases. Thus, the first change (T^ and T2) of primary 
adaptation is marked by an input of information describing the 
socialization demand of the larger system. This results in a general 
maladjusted period in which achievement is hampered, rewards 
are reduced, satisfaction is lowered and values are deviant. In the 
second period ( T2 to T3 ) no new information is presented, but 
situational rewards and narcissistic rewards are increased so that 
gratification and satisfaction is increased. As socialization begins 
to take effect ( T4 ) , achievement actually increases, new values are 
introduced and accepted, rewards are kept at a high level. Then, 
as the sub-system falls into line, rewards are reduced (the child 
is expected to perform in his new status without being constantly 
told how good he is to do so ) and the new values are tied into the 
already existing system, so that the child feels his new performances 
as obligations that are "right" rather than as instrumental actions 
for rewards. The effect of the entire cycle is to increase achieve- 
ment level through the incorporation of new information and to 
bring the child's values into closer conformity with those of the 

178 • Alfred L. Baldwin 

superordinate system. The net eflFect on rewards is zero. The loss 
of rewards during the early stages of the cycle is balanced by in- 
creased rewards as the child begins to meet the new demands 

This conceptualization of socialization is a magnificent eflFort. In 
detail it is not always clear and is surely oversimplified but its 
difiiculties should not blind us to the fact that socialization pres- 
sure on the child is not merely an antecedent variable whose conse- 
quence is personality change. The whole process is the behavior of 
a social system with various feedbacks and other patterns of inter- 
relationship. Parsons has certainly provided the most ambitious 
attempt yet made to encompass this complex phenomenon of family 
life, and has especially emphasized how each aspect of the process 
plays its functional role in the operation of the system. 


After this long period of attention to the oedipal phase of 
development, we return to the sequence of stages of personality 
development during school age and adolescence. In the first three 
stages of socialization, the child is devoloping a set of need-dis- 
positions that are not, in principle, limited to a single object or 
even a single class of objects. In other words, these early acquisi- 
tions are attitudinal patterns like love, esteem, and so on, that will 
recur throughout life in connection with various social objects. But 
in the nuclear family, Parsons says, each of these attitudes is 
focused on a single social object. It is only later that the child 
learns the modifications of the attitude appropriate for different 
types of objects. Because the young child's contacts are largely re- 
stricted to the nuclear family, no demands for object discrimina- 
tion within the same attitude are imposed. This artificially imposed 
correspondence between attitudes and objects provides the child 
with a simplified social environment. Parsons argues that it is 
essential for the child to develop the attitudes first, and that such 
an orderly arrangement both prevents confusion and establishes 
the need-dispositions very firmly. The psychological rationale of 
this last consequence is not clear. 


Beyond the oedipal stage, socialization involves the difiFerentiation 
of objects rather than attitudes. These differentiations involve, 
naturally, the two object-choice pattern variables, universalism-par- 
ticularism, and pei'f onnance -quality . These differentiations are 
brought about by the child's activities outside the home. The father, 
up to this point, is a particular man as far as the child is concerned. 
All the child's attitudes are particularistic. In the peer group the 
universalistic categories of men, women, boys, girls, adults, chil- 
dren are all developed and the need-dispositions modified appro- 
priately. Parsons suggests that universalistic values presuppose 
categorization on the basis of abstract properties, and points to the 
fact that logical thought develops during this same period of life. 

The final stage of differentiation involves the performance-quality 
pattern variable. It takes place particularly during adolescence and 
marks the emergence of the child into full adulthood. Heterosexual 
adjustment and meeting the occupational demands of the society 
are the components of the adolescent crisis that Parsons believes 
are mainly responsible for this differentiation. Although Parsons ex- 
plicitly states that the differentiations occur in this order, it would 
perhaps not be doing serious violence to the theory to suggest that 
the last two differentiations overlap considerably and perhaps even 
occur simultaneously. Certainly, much of the training in adaptive 
behavior that characterizes the latency period would require the 
discrimination between quality and performance types of relation- 

The result of these sequences of differentiations leads to what 
Parsons speaks of as a genealogy of need-dispositions as shown in 
Figure 10. Parsons sees this genealogy as representing a series of 
irrevocable bifurcations of motivation. The conceptual model 
adopted here is that of branching streams of motivational energy 
starting from an undifferentiated source essentially equivalent to 
Freudian libido. 

It is clear that these need-dispositions are concerned with many 
aspects of behavior and that the dispositions are of widely different 
types. On the one hand, there is a need-disposition for "pleasure in 
gratifying states." This seems general enough. There is also a quite 
specific need-disposition toward cooperation for team success. On 
the other hand, there are no need-dispositions describable as fear, 


.2 ^ 

(D 3 

P 3 



























■w (U 



OQ ^ 










u s 

3 O 

ar <u 

«J cd 


































w X3 

ft u 






















2 S 

CQ ^ 



<u ^ 

O n^ 

•S u 

o $ 



3 ^ 


1 O 

J, 3 

S ''' 


o 5 

n1 o 




cii A 

rt ,C 

2 o 






























^ M 












spaa^I oiubSjo 


avoidance, hostility, competition, seclusiveness, or possessiveness. 
Their absence in Figure 10 does not mean that Parsons is uncon- 
cerned with such matters, but their absence is a serious problem 
for the theory. Parsons speaks of anxiety in several places and at 
least once calls it a need-disposition (GTA). Alienation is a 
tendency that is obviously important. It is a type of deviance stem- 
ming from the first stage but its conceptual status seems very vague 
indeed. It does" seem very difficult to see how Parsons' picture of 
motivation in Figure 10 can fit the facts of child behavior. The pic- 
ture contains some interesting personality variables, but it is hard 
to escape the feeling that the bifurcation model is accepted be- 
cause it fits the theory rather than reality. 


Turning now to the relation between personality and the 
social system, we find that values and role-expectations are intro- 
duced in FSI not as need-dispositions but as integrating forces upon 
need-dispositions. In GTA, on the other hand, they were themselves 
need-dispositions. The personality is organized in two directions. 
One is based upon need-disposition and is responsible for temporal 
sequences of behavioral acts leading to goal states. The other is a 
series of "cross ties" ( Parsons' term ) corresponding to a set of value 
patterns common to other members of the social system. A quota- 
tion seems in order here: 

The essence of a system of action, then, is that it consists of 
motivational or need-disposition units each with its differentiated 
goals, interests and sentiments but bound together with other 
units by serving the interests of the same value patterns, each 
of which mobilizes a pluraliy of different motivational types or 
units. Seen in personality terms, these value systems are sti^ate- 
gically the most important properties of internalized social ob- 
jects. (FSI p. 167) 

Parsons believes that need-dispositions, even those involving 
cooperation for team success and discipline in pattern maintenence, 
will by themselves "fly off in all directions." "They must be held 

182 • Alfred L. Baldwin 

together as a team in the service of the system as a system." ( FSI ) 
The function of the common vahies is to achieve this integration. 

Now, however, a new notion is introduced; namely, "that a single 
pattern of values is not adequate, there must be a ramified system 
of such patterns, the structure of which matches the differentiations 
of structure of the relevant systems of action, both personality and 
social systems." (FSI) Roles are formulated at the points of inter- 
section of these values patterns. 

By this, Parsons apparently means that— in the social system at 
least— all those in the system have some common values. Further- 
more, all who have the same role share a unique set of values held 
by nobody who does not have this role. It would seem that perhaps 
all members of the system have value a. Some hold b. Some hold 
value c. Perhaps all the people who have a certain role have both 
values b and c and nobody else has that particular set of values. It 
is in this sense that roles correspond to "intersections" of values. 
Parsons does not make it clear whether every different set of values 
will correspond to some role. 

Now, in the personality system, this same set of values integrates 
the need-dispositions in such a way that the proper role behavior 
is performed. Parsons refers to roles in the family to illustrate the 
significance of this formulation. Thus a family has a value system 
shared by all of its members and defines its system goals and norms. 
The marriage relationship is a sub-system with values especially 
related to tension-management. The husband has a unique position 
in this relationship because, in addition to the values shared with 
the wife, he also holds values appropriate to the male role, to his 
occupational role, and so on, and this particular constellation of 
values uniquely characterizes the "husband" role. Since the male 
role is marked by more adaptiveness than the female role, and since 
the occupational system involves adaptive values more than the 
family system, the husband's role in the family is more adaptive 
and instrumental than the wife's. 

This formulation is very neat. It provides a system for describing 
the influence that participation in one sector of a social system has 
upon the role behavior in some other sector of the sub-system. It 
also provides a way of describing genuine uniqueness in role be- 
havior without ascribing it to idiosyncratic personality or setting 


up an idiosyncratic role. Each husband is the intersection of a some- 
what diflFerent set of values from every other husband. Thus, he 
fills the husband role in a somewhat different way. 

Now let us see if we can fit these concepts into a theory of per- 
sonality. Social objects are a focal point, not only in the socializa- 
tion process, but also in the structure of personality. The term, 
internalized social object, however, might have several meanings. 
In its simplest form it is merely a conception of the social object or 
a cognitive representation of the social object. This is the sense in 
which Olds uses the term in Structure and Growth of Motives where 
he is developing Parsonian conceptions. In the description of the 
oedipal transition (supra, pp. 172-76), the child's conception of the 
mother gradually changes, but there is nothing in that discussion 
to indicate that "internalized social object" has any further implica- 

On the other hand. Parsons obviously intends internalization of 
the social object to have somewhat the same meaning Freud gave 
it. It implies a taking over of the motives and values of another 
person, including apparently the values associated with each of the 
roles of that alter. In the personality of ego, therefore, it might be 
represented merely as a constellation of values associated wdth the 
particular roles learned from some other person. In other words, 
it would not carry around in the personality any label denoting 
just what social object it was that had been internalized. This in- 
terpretation seems to be strongly suggested by the quotation on 
p. 181 supra, "Seen in personality terms, these values systems are 
strategically the most important properties of internalized social 

As in so many other areas of Parsonian formulation of internali- 
zation, this contains some interesting ideas, but when examined in 
detail it becomes blurred. In all justice, however. Parsons' usage 
is no more confusing than most other discussions of the concept. 


In discussion of the Parsonian theory of personality de- 
velopment, we must avoid, if possible, several dilemmas. 

184 • Alfred L. Baldwin 

It is important to distinguish between a man's comments and his 
theories. With Parsons there are, almost two different people in- 
volved. His discussion of concrete issues, such as the role of the 
doctor in American society, or the general discussion of the Ameri- 
can family at the beginning of Family, Socialization and Interac- 
tion Process, is often clear, cogent, perceptive and exciting. When 
the theorist begins to talk, however, his style becomes more diffi- 
cult, his sentences awkward, and his meaning unclear. Theoretical 
language frequently sacrifices liveliness and grace for accuracy of 
communication, but it must be accurate. We have seen several ex- 
amples of how Parsons' theoretical language is far from precise. It 
is full of shifting meanings and vague antecedents. For a theory 
this is a fatal defect. Parsonian theory cannot be a good theory 
until rewritten in a coherent, consistent fashion. 

A conceptualization cannot be dismissed summarily, however, 
just because it is obscure— or for that matter, because it is wrong. 
We have a few serious attempts at general theory building in social 
science and no clear successes. Most attempts at theory in social 
science must be evaluated and criticized partly like literature and 
only partly like science. If man's conceptualization of human be- 
havior provides insights that enable others to understand it better 
and function more effectively, it has value even if it is an untenable 
scientific theory. But a social scientist, in contrast to a social com- 
mentator, must devise his conceptualizations with an eye to their 
incorporation into genuine theories and to their ultimately be- 
coming testable. Therefore, the critic of these proto-theories must 
try to be a seer and estimate the contribution a conceptualization is 
likely to make to the development of genuine scientific theory. 

With that ambitious purpose in mind, what might be said about 
the possible contributions of Parsons' concepts to the development 
of a theory of personality development? 

The functional viewpoint of the family: Parsons emphasizes over 
and over again the interrelations between the personality and the 
social system. He tries to trace out the effects of a personality 
change on the other people and various sub-systems within the 
family. Further, he tries to show how these effects reflect back on 
the individual as social pressures and stimuli, that result in further 
changes until the entire system reaches some reasonably stable 


state. This efiFort can hardly help but contribute to personality 
theory. It is a healthful counteraction to the tendency of psycholo- 
gists to view environment solely as cause of behavior rather than 
as eflFect also. Even if it fails, it is an important point of view that 
will be less likely to be ignored now that Parsons has emphasized 
it so strongly. 

The isomorphism between personality and social system: This 
conviction of Parsons seems to the present critic the root of many 
difficulties and confusions. It began with Bales' assumption of 
isomorphism between small face-to-face groups in the laboratory 
and the society as a whole. In that setting the assumption led to 
interesting research and novel methods of analysis. The empirical 
findings suggested immediately the possible value of drawing a 
parallel between the family and society. 

The isomorphism of the personality and the social system is, how- 
ever, a diflFerent matter. The effort to maintain this analogy seems 
responsible for some of the basic troubles with the theory. For ex- 
ample, the absence of so many needs in Figure 10 seems to stem 
isomehow from the insistence that every need-disposition must cor- 
respond to some social role, and every need-disposition must corre- 
spond to one combination of the five pattern variables. 

Perhaps the difficulty lies in a too rigid interpretation of the 
parallels between one dynamic system and another. It is not at 
all unlikely that both the personality and the social system are 
dynamic systems whose components are interdependent and whose 
pattern is in some sort of equilibrium, but such a faith does not 
require the assumption that need-dispositions correspond to people. 
The problem of integrating motives within the person bears only a 
slight resemblance to that of integrating people in a society, al- 
though the problem of integration exists in both systems. People 
are not caricatures of a particular motive— the glutton, the sadist, 
and so on. They have complicated behavior patterns in different 
situations and this complexity is involved in their integration into 
social systems. People are not conceptual units, they are chunks of 
flesh within a skin operating as a physical entity. Need-dispositions, 
on the other hand, are conceptual constructs defined to have only 
one property— motivation. To endow them with sufficient com- 
plexity to sustain the analogy to the person would destroy their 


186 • Alfred L. Baldwin 

usefulness as motivational elements. If one felt forced to draw 
an analogy between personality and the social system, it would 
seem much more sensible to make needs correspond to social roles 
rather than to individual actors. 

This argument, if valid, leads to the judgment that Parsons' in- 
sistence on isomorphism is not one of the valuable contributions of 
his work on which social science may build. 

The impoverishment of personality: It seems fair to say that Par- 
sons fails in his theory to provide the personality with any reason- 
able set of properties or mechanisms aside from need-dispositions, 
and gets himself into trouble by not endowing the personality with 
enough characteristics and enough different kinds of mechanisms 
for it to be able to function. 

Even when he is writing chapters on personality structure, Par- 
sons spends many more pages talking about social systems than he 
does about personality. He draws the analogy between the per- 
sonality and the social system strongly, but does not spend much 
time discussing the way individual people function. His descrip- 
tions of psychological reactions, when they do occur, frequently 
show insight but are couched in the everyday language of common 
sense. For example, in Family, Socialization and Interaction Proc- 
ess, he discusses the phase labeled "primary adaptation." This is 
a kind of jar or shock that leaves the child up in the air; he is 
frustrated. Parsons then lists, in the course of a few paragraphs, 
the reactions that the child makes— such as clinging to the old state 
of affairs, trying to get the mother to take the same kind of care 
that she previously did, responding to the frustration with aggres- 
sion, or displacing the aggression. Not all of these responses are 
really provided for in the theory. He describes what happens when 
people become frustrated, drawing upon common sense and 
Freudian theory, but he is merely describing psychological reac- 
tions, not putting them into his theory. 

In The Social System there is a very interesting discussion of the 
medical profession. In it Parsons points out the conflict that arises 
because a doctor who is devoted to helping the patient and is 
collectivity-oriented as far as the pair is concerned is restricted to 
dealing with a specific aspect of the patient's life and gets paid for 
doing so. The collectivity-orientation seems to Parsons to lead 


neutrally to a diffuse relationship between the doctor and patient. 
This diffuse relationship is not entirely compatible with specializa- 
tion o£ functions. The collectivity-orientation is similarly out of tune 
with receipt of payment. 

This observation is shrewd, but there is no psychological mecha- 
nism in the theory that would make it possible to derive the 
prediction that a collectivity-orientation tends to lead the individual 
into a diffuse relationship with the other person in the collectivity. 

A third example can be found in the discussion of oedipal sociali- 
zation. Parsons is trying to describe how the child has, to some ex- 
tent, lost the mother as an instrumental agent of care, and how he 
must shift over to the acceptance of the father as the instrumental 
person in the family. The father does not actually take over the 
mother's care functions, but, nevertheless, functions as a more in- 
strumental person than the mother. This description poses a very 
interesting psychological problem. If the child has perceived the 
mother as instrumental and now she stops being so, just how would 
he now come to perceive the father as instrumental? 

Parsons struggles with the problem, and suggests that the transi- 
tion occurs via the parental role which the mother and father share. 
Gradually, in this way, the child's perception of the instrumental 
function moves from the "mother" to the "parent" to the "father." 
This may or may not be a valid account, but even if it is, Parsons' 
theory, however persuasive, seems to me to provide no psycho- 
logical mechanism. When Parsons is describing psychological re- 
actions of individual people, he frequently proceeds on common- 
sense grounds, and not by way of any psychological theor)^ 

The importance of the psychological basis of behavior may be 
well illustrated by reference to roles. Parsons defines roles in terms 
of mutual expectations and tacitly assumes that these expectations 
are, in fact, the psychological instigators of role-behavior. As a 
result, he does not distinguish the role-definition from the psycho- 
logical mechanisms by which people are led, guided, instigated, 
forced, or rewarded into actually complying with the role-expecta- 
tion. People do fulfill role requirements, but they fulfill them in a 
variety of ways, and by way of a variety of psychological mecha- 
nisms. Sometimes they fulfill them as instrumental acts to avoid 
punishment or to gain rewards. Sometimes people fulfill the role 

188 • Alfred L. Baldwin 

requirements because they are motivated to conform; for such 
people, the presence of a standard, is suflBcient to instigate conform- 
ing behavior. Other people— or the same person at diflFerent times 
—may fulfill role requirements because they feel a moral imperative 
about the behavior itself; for these people it may be unimportant 
that a certain set of behaviors constitute a social role; for them the 
actual behavior itself is seen as "something I ought to do." Other 
people may fulfill role requirements through love, loyalty, or senti- 
ment for another person or for a collectivity. The mother cares for 
her child— and in this way fulfills the role requirements of the 
mother— largely through a natural expression of this love for the 
child. Role requirements may also be met because the actual role- 
behavior is itself a consummatory, rewarding kind of action. 

We see that there are many different psychological mechanisms 
that can underlie the same role-behavior. Parsons sometimes ap- 
pears to be thinking of one of these mechanisms and sometimes 
another, probably without really distinguishing between them. 

We should ask whether these psychological differences under- 
lying role-behavior make any difference to the social system, for 
Parsons is clearly primarily interested in talking about the latter. 
If the distinctions that the psychologist makes amount to no more 
than hair-splitting, Parsons is quite right to ignore them for his 

The distinctions are, however, important ones for the social sys- 
tem. As long as the social system is functioning smoothly, and people 
are, in fact, carrying out all their roles, as long as the roles are 
intermeshing and interdigitating properly and the functional re- 
quirements of the system are being met, the reasons why people 
perform their roles are unimportant; the important thing is that the 
actors do perform. 

But Parsons does not intend to restrict himself to the smoothly 
functioning social system. He is very much interested in strains and 
changes in the system. Here, the psychological basis for role-be- 
havior makes a big difference. Role-behavior that is merely instru- 
mental for getting a reward will break down quickly under cer- 
tain kinds of strain, and, on the other hand, can be easily changed 
to a different kind of role-behavior if the person finds himself in a 
new situation. Role-behavior, on the other hand, that is consum- 


matoiy in its own right is much harder to change. Its stabihty, 
therefore, could lead both to stability of a social system under 
strain, or to rigidity of the social system if change were necessary. 
When Parsons oversimplifies and impoverishes the personality in 
his theory, it can have serious consequences on the ability of the 
theory to handle the problems that Parsons himself feels are cen- 


Any evaluation of Parsons' contributions to personality 
theory would be incomplete without some attention to his central 
hypothesis that socialization is a double cycle in which the so- 
cializing agency goes through a task-performance cycle: adapta- 
tion, gratification, integration, and latency; whereas, the socialized 
system goes through a reverse cycle analogous to psychotherapy. 
The trouble with evaluation of this hypothesis is that it is funda- 
mentally an empirical one, but there are no data on which to judge 
it. In part, this particular hypothesis is testable; it could be investi- 
gated by recording, in some fashion, all the interactions in a family 
relevant to some particular aspect of socialization Uke toilet train- 
ing. The data should include parental discussions about toilet 
training, and other relevant interactions with each other, as well 
as with the child. These interactions could be coded by some 
scheme similar to Bales' interaction analysis and should show a 
sequential patterning as described by the theory. Although such a 
research project would be di£Bcult it would be feasible and en- 

To test empirically the progress of the child from two to four 
and then to eight degrees of differentiation is much more difficult. 
In fact, it is probably impossible without adding many assumptions 
and operational definitions to the scheme as proposed in Familij, 
Socialization, and Interaction Process. Some investigation of little 
pieces of the theory, however, might be possible. It would be in- 
teresting, for example, to see how the child's conception of the 
mother as reflected in his behavior toward her changes as she 
moves from a more to a less instrumental role. For example, the 

190 • Alfred L. Baldwin 

child will surely ask for less care in the areas where socialization 
has progressed, but as he meets new difficulties requiring help in 
new sorts of instrumental behavior, does he turn less to the mother 
and more to the father for instrumental help? 

As far as the main empirical content of Parsons' theory is con- 
cerned, therefore, the evaluation must await relevant data. It is 
reassuring to find that some kinds of relevant empirical data can be 
obtained. The theory is not completely untestable, but it is fair 
to say that many of its features cannot be tested without much 
clearer definitions of the relevant terms. 

In summary, it seems to this critic, that personality theory has 
quite understandably been somewhat of a sideline rather than a 
central aspect of Parsons' contributions. His sociological perspective 
has led him to emphasize very important features of personality de- 
velopment usually neglected by the psychologist, notably, the way 
a person's behavior has consequences on other people whose reac- 
tions eventually feed back to cause changes in the person. On the 
other hand, he has so impoverished the personality that it cannot 
function effectively, even in his theory and for his problems. This 
may be a result of his functioning viewpoint, but may also stem 
from what this critic feels is a basic difficulty, a rigid insistence 
upon isomorphism between the personality system and the social 
system. Despite these difficulties, the theory is certainly valuable 
and suggests important lines of research. If it does this, its defects 
can be quickly forgiven. 



Urie Bronfenbrenner 






It is a familiar tenet of most theories of identification 
that close association with another leads to the taking on 
of his characteristics. What is more, the theories assert 
that such assimilation takes place even— perhaps especially 
—when the particular other is an object of censure. We 
would suggest that this same phenomenon occurs in the 
realm of scholarly and scientific criticism; that is, the critic 
is likely to take on the characteristics of the person criti- 
cized—even those characteristics which he most vigorously 
assails. The more completely he strives to come to terms 

Note: This essay represents a restatement and further extension of 
ideas initially proposed in an earlier paper for presentation to the 
faculty seminar by the author. Cf. U. Bronfenbrenner, "Freudian 
Theories of Identification and Their Derivatives," Child Development, 
31:15-40 (March 1960). 

192 • Urie Bronfenbrenner 

with the specific substance and mode of thought of the author's 
work— even if only to attack them— the more his own conceptions 
are apt to approach a correspondence— if only in opposition— to 
the formulations he seeks to evaluate. We would submit further 
that Parsonian theory offers a case in point. This theory is often 
censured for its protean ambiguity, pretentiousness, lack of opera- 
tional referents, excessive preoccupation with overarching schemata 
to the neglect of constituent substructures, and failure adequately 
to relate the theory and distinguish it from the work of others. 
Yet paradoxically the major critiques of Parsons' contributions 
often suffer from these very shortcomings. In their efforts to do 
justice to Parsonian theory "writ large," they too are apt to rise 
to airy levels of abstraction, to lose sight of the hard ground of 
empirical fact, to perceive an over-simplified and incomplete view 
of the complex terrain below, and to be uninterested in the more 
delimited but detailed descriptions of those who have painstakingly 
surveyed segments of the same territory on foot. 

In this essay, we shall strive to resist such pressures to isomorphism 
by deliberately focusing our attention on a restricted segment of 
Parsons' thinking— a segment to which, for once, he does give con- 
siderable attention— and endeavoring to evaluate this delimited con- 
tribution in the light of theories developed by other workers to deal 
with the same types of phenomena. 

The theoretical segment selected for this comparative analysis 
is Parsons' extended treatment of the process of identification. Dis- 
cussions of this process appear in a number of sources (notably 
The Social System, Toward a General Theortj of Action, the Work- 
ing Papers, and Family, Socialization and Interaction Process), but 
the principal exposition of this aspect of his theory is set forth in 
the volume on the family. We shall draw on aU these sources in 
our review. 


Since Parsons admittedly takes Freud's theory of identifi- 
cation as the point of departure for his own formulations, we must 
first acquaint ourselves with the basic features of the psychoanalytic 

parsons' theory of identification • 193 

view. Having recently attempted to collate and integrate Freud's 
widely scattered writings on this topic,* the author has drawn 
principally on this secondary source for preparing the summary 
which follows. 

Unfortunately for our purposes, in his extensive discussions of 
identification, as of all other topics, Freud was no less prolific or 
protean than his sociological successor. In general, we can dis- 
tinguish three major uses of the term "identification" in his writ- 
ings. Most often, as in his discussion of the Oedipus complex, 
Freud treats identification as a process— the sequential interplay of 
forces internal and external which impel the child to take on the 
characteristics of the parent. 

But, on occasion, Freud also uses the term "indentification" to de- 
scribe the product or outcome of the process— the resultant similarity 
in the characteristics of the child and the model. Moreover, there 
is the further question of what aspects of the model are being 
emulated. At times, as in the example of the boy who identifies 
with a kitten by crawling about on all fours, refusing to eat at the 
table, and so on, it is the overt behavior of the model which is 
being adopted. In other instances, as when Freud speaks of mould- 
ing "one's ego after the fashion of one that has been taken as a 
model," identification would appear to include internalization of 
the motives as well as the overt behavior of another. Finally, in 
his later writings it is not the parent's ego with which the child 
identifies but his superego, his idealized standards for feeling and 
action. In short, there are three aspects of the parent upon which 
the child may pattern himself: the parent's oveii; behavior, his 
motives, or his aspirations for the child. 

Finally, in speaking of identification Freud frequently puts 
emphasis neither on antecedent nor consequent variables but on an 
intervening construct— the notion of a disposition or motive. f For 
example, consider the statement by Freud that comes closest to 
being a formal definition of what he meant by identification. 

* U. Bronfenbrenner, op. cit. 

f Cf. U. Bronfenbrenner and H. N. Ricciuti, "The Appraisal of Per- 
sonality Characteristics in Children," in P. H. Mussen, ed., Handbook 
of Research Methods in Child Development (New York: John Wiley 
and Sons, 1960). 

194 • Urie Bronfenbrenner 

It is easy to state in a formula the distinction between an 
identification with the fathei- and the choice of the father as an 
object. In the first case, one's father is what one would like to 
be, and in the second he is what one would like to have. The 
distinction, that is, depends upon whether the tie attaches to 
the subject or the object of the ego. The foraier is therefore 
already possible before any sexual object choice has been made. 
It is much more difficult to give a clear metapsychological repre- 
sentation of the distinction. We can only see that identification 
endeavours to mould a person's own ego after the fashion of 
one that has been taken as a "model." * 

It is important to note that Freud's concern is not with a highly 
specific imitative impulse to mimic one or another isolated piece of 
behavior. Rather, he is positing a generalized tendency on the part 
of ego to take on not merely discrete elements of the model, but 
of the total pattern. Moreover, as Freud sees it, this tendency is 
more than a mere readiness or passive susceptibility. On the con- 
trary, it is characterized by an emotional intensity reflecting motiva- 
tional forces of considerable power. These features of pattern and 
power are reflected in Freud's use, as virtual synonyms for identifi- 
cation, of such terms as introjection and incorporation— words which 
connote a total and somewhat desperate "swallowing whole" of the 
parent figure. 

Although Freud gives less explicit attention to the concept of 
identification as a dispositional construct than to the discussion of 
associated processes and products, the conclusion is inescapable 
that it is the motive to become like another that is the organizing 
focus of his concerns; that is, he is interested primarily in the nature 
and consequences of the processes that impel, or even compel, a 
child to take on the characteristics of another person. 

To return, then, to Freud's theories of process. Here we have 
distinguished two mechanisms which for a long time remained 
fused in Freud's thinking but are ultimately differentiated. The first 
of these mechanisms involves identification as a function of loss of 
love, the second as a function of fear of the aggressor. We shall 


S. Freud, Group Psychology and the Analysis of the Ego (London: 
Hogarth Press, 1948), pp. 62-63. 

parsons' theory of identification • 195 

refer to the former as anaclitic identification and to the latter by 
Anna Freud's classic phrase identification with the aggressor, or, 
more briefly, aggressive identification. 

The process of identification with the aggressor is most clearly 
explicated in Freud's theory of the development of the Oedipus 
complex in boys. To recapitulate the familiar thesis: the boy sees 
his father as an all-powerful rival for the mother's aflFection; since 
he cannot overcome this rival (for fear of castration) he attempts 
to cope with the overwhelming power by allying himself with it. 
In common parlance, "if you can't beat 'em, join 'em." 

The principle of anaclitic identification, while implicit in Freud's 
early writings, is not fully explicated until Freud, in his later years, 
takes up the problem of the development of the Oedipus complex 
in women. Since women are presumably already "castrated" they 
would have no reason to identify with a threatening aggressor. 
What factors, then, motivate the girl to identify with adult stand- 
ards and modes of behavior? Here is Freud's answer. 

Fear of castration is naturally not the only motive for repression; 
to start with, it has no place in the psychology of women; they 
have, of course, a castration complex, but they cannot have any 
fear of castration. In its place, for the other sex, is found fear 
of the loss of love, obviously a continuation of the fear of the 
infant at the breast when it misses its mother. You wiU under- 
stand what objective danger-situation is indicated by this kind 
of anxiety. If the mother is absent or has withdrawn her love 
from the child, it can no longer be certain that its needs will be 
satisfied, and may be exposed to the most painful feelings of 

Such, then, are the concepts and processes which Parsons avow- 
edly takes as the starting point of his own conceptions. Before turn- 
ing to the latter, however, we must take note of some modifications 
of Freud's theory of identification which turn out to be especially 
relevant for Parsons' formulations. 

* S. Freud, New Introductory Lectures in Psychoanalysis (New York: 
W. W. Norton, 1933), p. 121. 

196 • Urie Bronfenbrenner 

III. mowrer's "alternative view" 

I refer to the revision of Freud's theory of identification 
proposed by Mowrer,* who, while accepting the context and char- 
acter of the mechanisms proposed by Freud, takes issue with the 
latter's view that identification is more difficult for the girl than 
for the boy. In view of the similarity of Mowrer's position to that 
subsequently taken by Parsons, we quote the pertinent portions of 
Mowrer's exposition. 

. . . Because Freud assumed that object choice is primary and 
identification derived therefrom, he believed that the psycho- 
sexual development of boys is simpler than that of girls, since 
boys can at an early date take women as sex objects and retain 
them as such throughout life; but girls, Freud conjectured, hav- 
ing, like the boys, taken the mother as the first sex object, must 
later abandon this object choice in favor of men and assume 
instead an identification relationship with the mother and with 
women generally. The alternative hypothesis here suggested 
holds that the situation is the reverse. Because the infant's first 
experiences of care and affection are with the mother, we infer 
. . . that there will be a tendency for children of both sexes . . . 
to identify with the mother. This provides a path of develop- 
ment which the female child can follow indefinitely; but the 
male child must, in some way, abandon the mother as a personal 
model and shift his loyalties and ambitions to his father. Once 
the boy and the girl are securely aligned with the mother, and 
the father, respectively, the terms of their basic character struc- 
ture, then, as specific sexual needs arise, they can be handled 
along lines prescribed as correct and proper for members of 
their particular sex. 

However, we must not neglect to consider the question of 
how it is that the boy, whose primal identification is ordinarily 
with the mother— for example, mothers almost certainly play a 
greater role in their infant's learning to walk and talk than do 
their fathers— how it is that the boy eventually abandons the 

* O. H. Mowrer, "Identification: A Link between Learning Theory 
and Psychotherapy," in Learning Theory and Personality Dynamics 
(New York: Ronald Press, 1950), pp. 573-616; O. H. Mowrer, 
"Neurosis and Psychotherapy as Interpersonal Process: A Synposis," 
in O. H. Mowrer, ed., Psychotherapy: Theory and Research (New 
York: Ronald Press, 1953), pp. 69-94. 

parsons' theory of identification • 197 

mother as his personal guide and takes instead the father. Here 
we have few facts to guide us, but we may plausibly conjectine 
that the first identification which infants make with mother 
figures is undifferentiated. By tliis I mean that the small child 
probably first comes to perceive the mother, not as a woman 
who is distinct from men, but simply as a human being, differ- 
ent in no systematic way from other adult figures in the environ- 
ment. The personal characteristics which are acquired though 
identification with, or imitation of, the mother during this period 
are characteristics or accomplishments which are appropriate to 
all persons, male and female alike. It is only at a later stage, 
presumably, that the child becomes aware of the partition of 
mankind into two sexes; and it is then that the fatlier, who has 
played a somewhat subsidiary role up to this point, noiTnally 
comes foi'ward as the boy's special mentor, as his proctor, guide, 
and model in matters which will help tlie boy eventually to 
achieve full adult status in his society, not only as a human 
being, but also in the unique status of a man. This, we note, 
involves two things: (1) being a man in the sense of being 
honorable, reliable, industrious, skilful, courageous, and courte- 
ous; and (2) being a man in the sense of being masculine, i.e., 
sexually oriented toward members of the opposite sex.* 

In short, Mowrer suggests that the series of identifications through 
which the child passes involves the progressive differentiation of 
social objects first with respect to age, and then to sex. It is this 
notion of sequential differentiation (apparently arrived at quite in- 
dependently ) which becomes the core of Parsons' theory of identifi- 


We begin our consideration of Parsons' theory by noting 
points of convergence with his psychoanalytic predecessor. To be- 
gin with, the generalized motive to become like another stands, 
with Parsons as with Freud, at the core of the concept of identifi- 
cation. Thus, in Toward a General Theory of Action, the following 
distinction f is drawn between identification and imitation: 

* Mowrer, "Identification," pp. 607-8. 

f The same distinction is drawn in The Social System ( p. 211 ), 

198 • Urie Bronfenbrenner 

Two major mechanisms for the learning of patterns from social 
objects are imitation, which assumes only that alter provides a 
model for the specific pattern learned without being an object 
of generalized cathectic attachment; and identification, which 
implies that alter is the object of such an attachment and there- 
fore serves as a model not only with respect to a specific pattern 
in a specific context of learning but also as a model in a general- 
ized sense. Alter becomes, that is, a model for general orienta- 
tions, not merely for specific patterns. (GTA 129) 

Similarly, if we interpret it correctly, Parsons' theory of the 
antecedents of identification is a restatement, in still another and 
even more esoteric language, of the now familiar mechanism of 
withdrawal of love. He takes as his point of departure the four-phase 
"paradigm of social control" originally proposed in The Social Sys- 
tem (300-26) and further elaborated in the Working Papers 
(238-45). Applying his conceptual scheme primarily to the thera- 
peutic process, Parsons distinguishes four sequential stages: per- 
missiveness ("allowing the patient to express himself"), support 
("to tolerate the excessive demands of the patient and 'accept' him 
as a human being"), denial of reciprocity ("the denial of response- 
reward, including . . . gratification in being duly punished for an 
aggressive act"), and manipulation of rewards ("a process of rein- 
forcing reality oriented' adaptive instrumental performance"). In 
the volume on the family (FSI), Parsons offers this same four- 
phase sequence as the basic foundation for his theory of socializa- 
tion and, more specifically, identification. Thus, in her treatment of 
the child, the mother begins with permissiveness and support which 
develop in the child "a diffuse attachment to her, a dependency on 
her." [italics Parsons'] 

We may presume that once dependency in this sense has 
come to be well established, the demand for attention, and for 
specific acts of care expands. The child manifests what, from the 
point of view of the mother's standards of child care, are illegiti- 
mate positive wishes. He is waked up at certain times though 
he would rather be allowed to sleep, he is given only so much 
to eat, less than he wants, he is put down when he would like 
her to continue to fondle him, etc. Whereas his dependency in 
general is welcomed and rewarded, excessive manifestations are 
pruned off by denial of reciprocity. . . . The balance between 
denial of reciprocity and positive reward gradually leads to the 

parsons' theory of identification • 199 

establishment of a stable "orientation" or expectation system in 
the child, the organization of his behavior both around the rela- 
tion to the mother as an object and involving certain standards 
of what are and are not legitimate expectations of his own 
gratification and of her behavior. When this process has reached 
a certain stage we can speak of the internalization of the mother 
as an object as having taken place. . . . This internalization is 
what Freud meant by ego's primary identification. (FSI 65) 

A more detailed analysis of the above sequence, this time in five 
stages, appears in the chapter by Parsons and Olds on "The Mecha- 
nisms of Personality Functioning." Here the parallelism with Freud's 
anaclitic theory is revealed in terminology as well as thought. In 
describing the process of internalization, the authors quote Olds' 
statement that the "ego must, after his initial shock 'return to' the 
old object." They go on to say: 

The mechanisms by which this occurs have to do with what he 
has elsewhere called the "law of motive growth" whereby, after 
being deprived in certain respects of gratification through an 
object, one comes to want it more intensely and more "uncondi- 
tionaUy" than before. (FSI 210) 

The new fifth stage of the sequence is identification itself. 

. . . the end product of this phase of the socialization cycle 
seems to us to be the appropriate place to use the term iden- 
tification. This essentially means that internalization of the new 
object system has been successfully completed . . . that from 
now on ego's major "predispositions" or "orientations" are to act 
in terms of the newly internalized object system and the motives 
which are organized in it. (FSI 229) 


Where then does Parsons diverge from Freud? Principally 
on the question of content, of what is internalized. The specific 
issues are raised in Parsons' essay on the superego (WP 13-29). 
The sociologist differs with the psychoanalyst on three major 
counts: First, Parsons criticizes Freud for failing to recognize that 
identification results in the internalization not only of moral stand- 

200 • Urie Bronfenbrenner 

ards (the superego) but also the cognitive and expressive features 
of the parent and through him of the culture as a whole. 

The general purport of this criticism is that Freud, with his 
formulation of the concept of superego, made only a beginning 
at an analysis of the role of common culture in personality. The 
structure of his theoretical scheme prevented him from seeing 
the possibilities for extending the same fundamental analysis 
from the internalization of moral standards— which he applied to 
the superego— to the internalization of the cognitive frame of 
reference for interpersonal relations and for the common system 
of expressive symbolism; and similarly it prevented him from 
seeing the extent to which these three elements of the common 
culture are integrated with each other. (WP 20-21) 

In a second and derivative challenge, Parsons takes exception 
to what he regards as an exclusively constitutional basis for Freud's 
theory of sexuality. 

. . . Freud speaks of the original "bi-sexuality" of the child. The 
presumption is that he postulated a constitutionally given duality 
of orientation. In terms of the present approach, there is at least 
an alternative hypothesis possible which should be explored. 
This hypothesis is that some of the principal facts which Freud 
interpreted as manifestations of constitutional bisexuality can 
be explained by the fact that the categorization of human per- 
sons—including the actor's categorization of himself as a point 
of reference— into two sexes is not, except in its somatic points 
of reference, biologically given but, in psychological significance, 
must be learned by the child. It is fundamental that children of 
both sexes start life with essentially the same relation to the 
mother, a fact on which Freud himself rightly laid great stress. 
It may then be suggested that the process by which the boy 
learns to differentiate himself in terms of sex from the mother 
and in this sense to "identify" with the father, while the girl 
learns to identify with the mother, is a learning process. One 
major part of the process of growing up is the internalization of 
one's own sex role as part of the self-image. It may weU be that 
this way of looking at the process will have the advantage of 
.making the assumption of constitutional bisexuality at least 
partly superfluous as an explanation of the individual's sex 
identification. In any case it has the great advantage of linking 
the determination of sex categorization directly with the role 
structure of the social system in a theoretical as well as an 
empirical sense. (WP 21-22) 

parsons' theory of identification • 201 

In the light of our own analysis of Freud's theories of identifi- 
cation, we are inclined to doubt Parsons' contention that Freud 
overlooked the possibility of learning as a mechanism in the de- 
velopment of identification. But by now the reader is in a position 
to judge the merits of the argument for himself. Parsons' final con- 
tention in the above quotation, however, can hardly be challenged. 
Certainly Freud has not linked "the determination of sex categori- 
zation directly with the role structure of the social system." Par- 
sons' third and major criticism of Freud focuses around this very 
issue and becomes the major theme of the sociologist's complex 
revision and extension of Freudian theory in Family, Socialization 
and Interaction Process. Here Parsons states: 

Freud was clearly very much on the right track, and in fact 
gave us the foundations of the present view. But what Freud 
lacked was a systematic analysis of the structure of social rela- 
tionships in which the process of socialization takes place. It is 
this which we are attempting to supply. (FSI 104) 

A prolonged eflFort on the part of this writer to extract the ele- 
ments of the Parsonian analysis convinces him that the word "sys- 
tematic" in the above quotation is being used in a truly Pickwickian 
sense. Adding to the usual difficulties of Parsonian prose (indeed it 
must be parsed to be understood) is the fact that the theory is 
stated at considerable length not once but twice— in Chapters II 
and III by Parsons alone and in Chapter IV by Parsons and Olds 
jointly, employing a somewhat diflFerent set of concepts. 

The basic features of the two formulations, however, are highly 
similar. The fundamental notion is that the child passes through 
not one but a series of identifications. The nature of these successive 
identifications is determined by the reciprocal roles being taken by 
parent and child at successive stages of the child's development. To 
understand these stages, however, we must first take cognizance of 
the four basic "status-roles" which Parsons regards as inherent in 
the structure of the nuclear family. These family roles are dis- 
tinguished in terms of two major axes, "symbolized," according to 
Parsons, "by the two great differentiations of generation . . . and 
sex." The first axis is that of power, with the parents being superior 
and the child inferior. The second is the familiar Parsonian polarity 


202 • Urie Bronfenbrenner 

of expressive vs. instrumental function. The former, associated 
primarily with the mother, involves being "affectionate, solicitous, 
warm, emotional to the children" and serving as "the mediator and 
conciliator of the family." In contrast, the instrumental function 
refers primarily to "establishing the desired relations to external 
goal objects" ( e.g., working at a job ) and acting as "the final judge 
and executor of punishment, discipline, and control over the chil- 
dren of the family." Given this last illustration of the instrumental 
orientation, the writer has diflSculty in distinguishing it from the 
presumably orthogonal factor of power. Be that as it may. Parsons' 
distinction between parental roles seems to parallel fairly closely 
Freud's descriptions of the nurturing mother and the punitive 
father. But one important difference may be discerned. The father's 
role, while involving discipline and control, is not predominantly 
hostile; it is also, and perhaps primarily, adaptive and directed at 
manipulation of the environment. We shall consider the implica- 
tions of this difference later in our discussion. At the moment, we 
return to the problem of developmental stages. 

Parsons emphasized that, at the outset, the young child cannot 
respond to the parental roles in their fully differentiated form. 
Moreover, the parental behaviors to which the child is exposed are 
segmental and not representative of the full role-repertoire of the 
parent. For example, in the beginning the mother's function is 
primarily instrumental; she gives the child physical care. Since 
care is not always forthcoming, this "denial of reciprocity" leads 
the child to his first identification— "the internalization of the mother 
. . . in her role as a source of care." Parsons emphasizes that "It is 
not the mother as a total personality as seen by adults that has been 
internalized, but that aspect of her with which ego has stood in a 
meaningful relationship of interaction." (FSI 65) Moreover, since 
for Parsons a role always implies a reciprocal relationship, the 
denial of reciprocity leads to an identification not only with the 
mother as an agent of care but also with a primordial image of 
the child himself as "the object of care." 

Herein lies the crux of Parsons' theory of the content of identifica- 
tion. At any given stage, the child identifies not with the parent as 
a total person but with the reciprocal role-relationship that is 
functional for the child at a particular time. Parsons stipulates a 

parsons' theory of identification • 203 

specific sequence of such role-relationships. Following his identifica- 
tion with the mother as a source of care, the child enters the stage 
of "love dependency" in which the mother's expressions of afiFection 
become rewarding in and of themselves; in other words, the child 
becomes responsive to the "expressive" aspects of the mother's func- 
tion. Since "a mother's love ... is always conditional," the denial 
of reciprocity at this level leads to internalization of the mother 
as a giver of love and himself as a loved object. 

Identification at the third or Oedipal stage reaches a new level 
of complexity. It is important to recognize. Parsons asserts, that at 
both earlier levels of identification the mother is still imdiffer- 
entiated with respect to sex. It is only in the Oedipal phase that 
the child first recognizes and internalizes the distinction between 
male and female, again simultaneously both in relation to his 
parents and himself. 

... the crucial event of this phase is the first stage of the as- 
sumption by the child of his sex role. The pre-oedipal child is, 
we assume, in the sense of fundamentally personality constitu- 
tion, sexless— as is in literal terms the "mother," since we assume 
that for the child the difi^erentiation of the two parents as ob- 
jects by sex has not yet on the requisite level been internalized. 
... In the earlier phases there was only one ascribed role the 
child could assume— more or less satisfactorily. Now he must 
"choose" between two— though the pressure to choose the ascrip- 
tively right one is overwhelmingly great. (FSI 78) 

Once more the differentiation occurs because of a shift in the 
parental role pattern presented to the child. Specifically, the expres- 
sive and instrumental functions are now divided between the 
parents with the mother specializing in the former and the father 
in the latter. 

... in the mother-child system, it was the mother who played 
the predominantly instnamental role, whereas in the wider family 
system of which the mother-child is, it will be remembered, a 
sub-system, it is the father. . . . This is to say that the father is, 
symbolically at least, the primary source of the new "demands" 
for conformity and autonomous performance. The mother, on 
the other hand, this time as distinct "person," remains the pri- 
mary source of "security" or "acceptance" in the love-relation- 
ship. (FSI 79-80) 

204 • Urie Bronfenbrenner 

As before, the pressure to discriminate and internalize the new 
role pattern is suppHed by denial of reciprocity and manipulation 
of reward, but now in line with the expressive-instrumental dichot- 
omy, these functions are distributed differentially between the two 
parents. The precise nature of this distribution and the manner 
in which it functions to produce differential identification for the 
boy and the girl is not fully clear from Parsons' exposition. On the 
one hand, in accordance with the expressive-instrumental distinc- 
tion, he allocates the first two stages of the four-phase paradigm 
of social control to the mother, and the last two to the father. 

Permissiveness and support, then, tend to be focussed on the 
mother role in the form of continuing nurturant care, and expres- 
sion of love. The more disciplinary aspect, however, focuses on 
the fathsr role, above all the denial of reciprocity and the manip- 
ulation of positive rewards for adequate performance. (FSI 80) 

Yet, on the very next page, the reward functions are allocated 
to the mother. 

Before he has internalized the father as an object the child 
cannot be fuUy sensitive to his attitudes as sanctions. He can, 
however, be motivated to do things which please both mother 
and father and be rewarded by mother's love and nurturance. 
By some such process he comes to cathect the father— because 
mother loves father and backs him up— and from the generalized 
parental object then a qualitatively different object can be dff- 
ferentiated out. (FSI 81) 

Although the division of expressive-instrumental functions 
between mother and father provides the child with a basis for 
differentiating the parents with respect to sex, it does not account 
for the selection of the appropriate sex role by the child. Having 
minimized the influence of constitutional factors, Parsons relies for 
an explanation on what is, in effect, a theory of differential rein- 
forcement for the two sexes. He states, "The new 'demands' of 
course this time are differentiated by sex. They consist in the 
appropriate forms of behavior for a l^ig boy' and a 'big girl' 
respectively." (FSI 80) 

As Parsons describes it, identification at the Oedipal level is 

parsons' theory of roENTIFlCATION • 205 

analogous to but nevertheless diflFerent from its manifestations at 
earlier stages. First, although the point is not made explicit, we may 
note that while at earlier levels the child was described as inter- 
nalizing the parent's overt behavior, at the Oedipal stage he is 
seen as motivated to identify not so much with what the parent 
himself does as what the latter regards as "appropriate." Implicit 
in this shift over successive developmental levels is an answer to 
the hitherto unresolved problem of whether identification refers 
to the parent's actual behavior or to his aspirations for the child. 
Parsons would seem to be suggesting that both are involved, but 
at di£Ferent stages of development. At an early age level, the child 
is able to identify only with the most concrete actions of the parent 
directly relevant to the child's well-being, and even then the 
identification is of a diffuse and relatively undifferentiated type. 
Later, with increasing capacity to abstract and discriminate, the 
child becomes capable of internalizing patterns which are at once 
more subtle and symbolic. Accordingly, if the parent "denies reci- 
procity" and "manipulates rewards" in relation to the symbolic 
aspects of the parent-child relationship (such as the child's con- 
formity with parental standards) then it is this more abstract 
reciprocal role pattern which becomes internalized. 

To put it in another way. Parsons' successive levels of identifica- 
tion represent a progressive differentiation of ever more complex 
role-relationships between the self, parent, and, ultimately, society. 
As in classical developmental psychology, the earlier differentia- 
tions are more diffuse, less stable— the later more specific and 
enduring. It follows that even though the same objective person 
serves as the model, the mother with whom, say, the girl identifies 
in infancy is quite different, both in her formal and substantive 
properties, from the mother internalized as an outcome of the 
Oedipal conflict. It was Freud's failure to recognize this distinction. 
Parsons asserts, which accounts for much of the ambiguity in the 
former's theory of identification. 

Returning to problems closer to the field of socialization, we 
may next raise again some of the questions involved in the con- 
cept of "identification." This has of course been a notably am- 
biguous and controversial concept in the literature of the field. 
Perhaps we can suggest some of the sources of the difficulties 

206 • Urie Bronfenbrenner 

and a constructive way out of them. Freud, it will be remem- 
bered, introduced the concept in connection with what, above, 
we have called the "mother-child identity." The principal diffi- 
culty seems to arise from the attempt to use the same concept in 
relation to the processes which go on in the oedipal period, 
above all with reference to sex-role assumption. Thus a boy is 
often said to "identify" with his father at this period and a girl 
with her mother. 

In our opinion the trouble comes from sticking to the attempt 
to deal only with the relation to one role-personality in a situa- 
tion where multiple role-relations are already involved. In the 
case of "primary" identification there was only one object, the 
nurturing or "caring" mother. Identification with this object 
could be treated as an adequate focus of the total internalization 
process. From the child's point of view, in the significant sense, 
he and the mother become one. 

When it comes to the oedipal period, on the other hand, for 
the boy his father is only one of four basic types of object. . . . 

What happens, then, is the reorganization of the total per- 
sonality as a system. This involves the addition, by fission, of 
two new object-units, the father as discriminated from mother 
and the discrimination of ego from sibling of opposite sex. There 
is also a differentiation of the collectivity structure from the 
simple mother-child "we" to a familial "we" with six potential 
elementary sub-collectivities. The focus of a son's identification 
with his father is his self-categorization as belonging to the 
"we-males" sub-collectivity— which is the same thing as saying 
that he and father share the category of "maleness." It means 
that this we is set over against a "they" of the females to which 
he cannot belong. But there is another they to which he also 
cannot belong, namely that of the "parents"— in this nuclear 
family— and his father does belong to this one. In this sense the 
boy cannot identify with, i.e., play the role of, his father, but 
only with his brother if any, and with respect to generation, not 
sex, his sister. It is, however, profoundly true that in this process 
both boy and girl internalize the father as an object. This aspect 
is strictly parallel with internalization of the mother in the 
primary identification, but the others clearly are not parallel, for 
the simple reason that in a one-unit system there are no analo- 
gies to many features of a four-unit system. . . . 

We suggest that the term identification has tended to be used 
to designate a variety of these different aspects of the total 
complex, but that the complex as a whole has not been ade- 
quately analyzed. We can suggest a usage of the term which is 
free of ambiguity, namely that identification should designate 

parsons' theory of identification • 207 

the process of internalization of any common collective "we- 
categorization" and with it the common values of the requisite 
collectivity. In this meaning of the teiTn, in the oedipal phase 
of development a child undergoes not one but three new iden- 
tifications. Two of them are common to members of both sexes, 
namely internalization of the familial we-category, and of the 
sibling category, namely "we children." The third, by sex, differs 
for children of each sex, in this third sense the boy identifies 
with his father, the girl with her mother. (FSI 91-93) 

As the reader may have already recognized. Parsons' reformula- 
tion of Freud's theory of identification is remarkably similar to the 
previously cited revision of Mowrer. The parallelism is even more 
clearly apparent in Parsons' views on the differences between the 
sexes in the resolution of the Oedipus complex. Like Mowrer, 
he argues that Freud was wrong in believing that the resolution 
is more difficult for the girl than for the boy. The latter, Parsons 

. . . has to undergo at this stage a double "emancipation." In 
common with his sister he has to recognize that, in a sense not 
previously so important, he must not pretend to adulthood, he 
is unequivocally a child. But as differentiated from her, he must 
substitute a new identification with an unfamiliar and in a very 
important sense threatening object, the father, at the expense of 
his previous solidarity with his mother. He must renounce his 
previous dependency in a more radical sense. The girl, on the 
other hand, though she must internalize the father as an object, 
does so only in his role as instrumental leader of the family as 
a system, not in the dual role which includes sex-role-model as 
well. Similarly, she remains categorized with her mother by sex, 
which coincides with the previous a-sexual (but not non-erotic) 
mother-child solidarity. Put a little differently, the boy must 
proceed farther and more radically on the path away from 
expressive primacy toward instrumental primacy. He is, there- 
fore, subjected to greater strain. (WP 98-9) 

Parsons goes beyond Mowrer in one important aspect. He points 
out that the child in identifying with the parent of the same sex 
begins to exhibit behavior which is sex-typed but bv no means 
identical with behavior of the adult parent. The discrepancy. Par- 
sons asserts, is more marked for the boy, and is related to two 

208 • Urie Bronfenbrenner 

factors— the clarity of the role model and the degree of anxiety 
generated by the very conflict which motivates the child to seek 
a new identity. 

The boy . . . tends to attempt to act out what are symbolic 
representatives of the instrumental aspects of adult masculine 
roles. These are notably nonfamilial in content. He plays with 
trains, cars, airplanes. He more or less explicitly assumes rela- 
tively tangible adult masculine roles such as fireman or soldier. 
He puts great emphasis on physical prowess. But his play is a 
less exact copy of the specific father role than his sister's is of 
the mother. This may well be explained, partly at least, by two 
facts. First the mother role is far more uniform than the mascu- 
line occupational role; the girl has a rather specific role-model 
stereotype. Secondly, being, as we have suggested, under less 
acute strain, the girl is less driven to the kinds of symbols which 
tangibly express compulsively tinged sex-qualities. Thus both 
the difficulty of understanding many middle-class occupations— 
their remoteness, and the fact that not involving physical 
prowess or skills, they do not patently symbolize masculinity- 
may prevent the urban middle-class boy from so directly emu- 
lating his father as the girl does her mother. (FSI 100) 

Here Parsons' argument is reminiscent of Freud's statement that 
the child identifies not with the parent as he actually is but with 
the parental imago, an image distorted in part by the child's own 
anxieties and needs. 

One final feature of Parsons' conception of identification at the 
Oedipal stage remains to be noted. In accordance with his theory 
of "binary fission," Parsons takes the position that the shift of the 
child from a two-member to a four-member social system is re- 
flected in a parallel development in personality organization. In 
his words, "the internalized object aspect ego differentiates from 
a two-unit personality structure to a four-unit structure by a process 
of bifurcation, on the instrumental-expressive dimension." ( FSI 78 ) 
The rationale of Parsons' next step is not fully clear. In a paragraph 
which could stand as a model of theoretical ellipsis, he writes: 

Turning to the need-disposition aspect of the new organiza- 
tion, we have treated the "dependency" need of the earlier 
stage, corresponding to the parental object, as divided into the 
"nurturance" need and the "conformity" need, as aspects of the 

parsons' theory of identification • 209 

internalized mother and fadier objects respectively. Correspond- 
ingly the "autonomy" need-disposition of the earlier phase is 
treated as dividing into those of "security" as the expressively 
differentiated or "feminine" self, and "adequacy" as the instru- 
men tally differentiated or "masculine" self -object. (FSI 83) 

Parsons does provide definitions— albeit somewhat idiosyncratic 
ones— of the four personality variables here introduced. Nurturance 
is "the positive gratificatory aspect of the original giving of 'care.' " 
Security is "the need to receive love or acceptance and to 'show 
solidarity' in relation to alter." Adequacy "refers to the autonomous 
performance aspect, the need and disposition to do specific things 
which are expected and acceptable." Finally, conformitif "refers to 
the need-disposition to enforce or to implement conformity with 
the highest level of normative standards which have yet been 
internalized." Parsons adds that "the superego clearly comes very 
close to what we have called the conformity need-disposition, and 
the internalized father-object clearly agrees with Freud." 

Granting that in a very general way these two pairs of disposi- 
tions reflect the giving and receiving aspects of Parsons' expressive 
and instrumental functions, it is nevertheless difficult to grasp how 
these particular personality characteristics become the necessary 
sex-specific products of the resolution of the Oedipal conflict. The 
origins of the presumed "masculine" attributes of adequacy and 
conformity are especially puzzling in this respect, i.e., how does 
it follow, from Parsons' theory as distinguished from Freud's, that 
the superego is the product of identification with the father and 
can be so "clearly" equated with the internalized recipient aspect 
of the instrumental function? This writer's effort to fathom the 
alleged inevitability of these theoretical interconnections was not 

Parsons' discussion of levels of identification beyond the Oedipal 
stage becomes increasingly recondite. The next social objects which 
become relevant for the child, he asserts, lie outside the family 
in the school and peer group. It is interaction in these social systems, 
Parsons argues, that enables the youngster to differentiate roles 
within a given category of age and sex. He states: 

. . . To an adult, a boy's father is of course only one instance of 
the universalistically defined category of "man." But to the 

210 • Urie Bronfenbrenner 

oedipal boy the discrimination of "father" and "man" has not 
yet been made, similarly, of "mother" and "woman," "self- 
brother" and "boy," "sister" and "girl." The question, then, is 
what social structures present these discriminations to him. . . . 

It is manifestly clear that the family cannot perform these 
functions by itself, because it does not have the necessary struc- 
tural differentiation. We have suggested . . . that family, school, 
and peer group should, in our society, for this purpose be treated 
as a single social system, comprising the whole range of the 
pre-adolescent's significant social participations. (WP 114) 

The nature of the role relationships in which the child becomes 
involved in school and peer group are not fully detailed. We are 
told only that "the instrumental subtypes are found mainly in the 
school, the expressive ones mainly in the peer group." (WP 52) 
Nor are the ways in which "denial of reciprocity" and "manipula- 
tion of rewards" operate to induce identification at this level clearly 
outlined; one is merely left with the impression that parents, 
teachers, and peers are all relevant as agents of socialization. 

The discussion of identification at the final level of the "adult 
community structure" is even more elliptical. Parsons' theory of 
personality development in adulthood reaches its climax in a table 
showing the proliferation of need-dispositions by binary fission to 
the level of 2* and a total of thirty-one variables comprising the 
"motivational complex of the mature personality system." One 
cannot help wondering what wondrous entity emerges from the 
next unspoken stage of binary fission, or the one after that? Perhaps, 
like its physical counterpart, the process reaches some crucial 
maximal level at which the entire structure explodes in a mush- 
room cloud of social chaos. 


At any event, we have completed our survey of a delimited 
aspect of Parsonian theory to which he accords considerable atten- 
tion and importance. How are we to evaluate this particular con- 
ceptual contribution? Having labored long and loyally in pursuit 
of new and useful theoretical ideas in Parsons' treatment of the 

parsons' theory of identification • 211 

topic, this reviewer has been disappointed. By and large, it has 
been difficult to discern much that is fundamentally new beyond 
the terminology. Although Parsons points emphatically to the need 
for revising and expanding Freud's theory in several directions, his 
own efforts along these lines fall far short of the expectations he 
creates; either he restates in even less precise language ideas that 
are already familiar in the writings of others or he offers concep- 
tions which, though provocative, are so diffuse that the basic tasks 
of theory construction have to be performed by the reader himself. 
Accordingly, it is perhaps merely a reflection of this author's limita- 
tions as a theorist that after earnest and repeated perusal of Parsons' 
writings on identification, he has had little success in deriving for- 
mulations that substantially modify or clarify existing theories in 
this sphere. Specifically only two ideas emerged from this analysis 
which, to the writer's knowledge, are not found in earHer treatments 
of identification. Both of these derive from Parsons' distinction 
between instrumental and expressive functions and the differential 
allocation of these two functions between the parents. If, as Par- 
sons suggests, withdrawal of love can be used as a technique for 
motivating the child to identify with the instrumental pattern of 
moving out into and manipulating the environment, then we have 
a possible basis for positing different processes and products of 
superego formation in the two sexes. Thus male morality may be 
more concerned \\ath general principles of conduct in relation to 
the outside world, while female morality centers about personal 
feelings and intimate interpersonal relationships. Such variation 
in the content or focus of the superego could be produced through 
the use of withdrawal of love in two somewhat different contexts. 
With girls this technique might be employed principally with 
reference to intimate family relationships, with boys more in regard 
to performance and achievement both within and outside the home. 
Finally, since the fathers, who are presumably instrumentally and 
externally oriented, are likely to have closer associations with sons 
than \vith daughters, boys would be more likely to develop a 
principled and objective (situation-oriented) superego. In con- 
trast, girls— who, according to recent research findings,* are par- 

* R. R. Sears, E. Maccoby, and H. Levin, Patterns of Child Rearing 
(Evanston, Illinois: Row, Peterson, 1957). 

212 • Urie Bronfenbrenner 

ticularly likely to be subjected to withdrawal of love by the mother 
—would tend to develop a superego highly susceptible to guilt over 
interpersonal matters rather than broader issues. 

Parsons' picture o£ the father as executive rather than punitive 
suggests also a variant of Freud's theory of identification with the 
aggressor. The fact that the father exercises power and control 
over the environment may in itself invite emulation. To the extent 
that an exploratory or activity drive exists (perhaps differentially 
for the two sexes ) , a living example of patterns of action for express- 
ing this drive may be sufficient to motivate the child to adopt an 
analogous pattern in his own behavior. Given the possibility of 
such a mechanism, a child whose father or mother was especially 
active in the manipulation of the environment, either through direct 
activity, or through the exercise of power (e.g., making plans, 
decisions, and so on) might be expected to emulate the parent's 
behavior, even without reinforcement through "denial of reciproc- 
ity" and "manipulation of rewards." 

But we have wandered far from the Parsonian orthodoxy, if such 
there be, and must return to our consideration of his own explicit 
views. Staying at the very general level at which he has chosen to 
couch his formulations, we may summarize the major tenets of 
his position as follows: 

First, Parsons makes explicit the thesis that the type of identifi- 
cation is a function of the developmental capacities of the child. 
Early identifications are more diffuse and related to concrete behav- 
ior; later ones are more differentiated and organized around sym- 
bolic role-entities. Second, Parsons stresses repeatedly that identi- 
fication involves not only motivational but also cognitive elements; 
the cathected pattern is determined in part by the substantive and 
formal properties of the model. To put it in another way, what 
the child strives to internalize will vary with the content and 
clarity of the reciprocal role relationship in which he is a participant. 

One can hardly quarrel with propositions such as these, and this 
is the most disappointing fact about them. In attempting to extract 
the essence of Parsons' thinking on a specific problem, we have 
emerged with a set of formulations which, while they could have 
been and were interesting and even exciting a generation ago, are 
now commonplace in developmental and social psychology. We can 

parsons' theory of identification • 213 

find these same ideas not only in Freud, but also— often expressed 
in more comprehensible and rigorous form— in the writings of such 
diverse theorists as G. H. Mead, Piaget, Sullivan, Werner, Cottrell, 
Heider, and Newcomb. One may counter that we have done Par- 
sons an injustice by reducing his ideas to everyday English and 
failing to take account of the connotative meanings of his termi- 
nology. Some years ago, in a mood of naive optimism, the present 
author argued that to be fruitful a theory "need not be stated in 
testable or even communicable form;" * having since savored of 
the fruit, he now regretfully confesses that it is neither tasty nor 

* U. Bronfenbrenner, "Toward an Integrated Theory of Personality," 
in R. R. Blake and G. V. Ramsey, eds., Perception: An Approach to 
Personality (New York: Ronald Press, 1951), pp. 206-257. 



Henry A. Landsberger 





The Relevance of Organization Theory to Parsons 

For Parsons, complex organizations are an excellent test 
of his general theory and, if his theory is correct, organiza- 
tion theorists should in their turn have much to gain from 
it. Parsons' definition of organizations makes them appear 
to epitomize the three problems faced by social systems 
in which he has been most interested: problems which are 
known as the dilemma of freedom versus order in the 
elegant but cryptic language of social philosophy. 

First, formal organizations contain subunits (individuals, 
departments and functions, occupational groups), and 
organizations can in turn be thought of as subunits of 
larger systems (such as the educational system or the econ- 
omy). Through what mechanisms and with what success 
can the activities of units at one level be integrated into 

parsons' theory of organizations • 215 

a higher level? To what extent should they and need they be 
so integrated? Students of organizations have paid a great deal 
of attention to these problems. Those studying these problems at 
one organizational level have often not been aware, and might 
not concede, that they have parallels at other levels within the 
organization and beyond it, let alone that organizations may share 
the problem with other parts of society. Making such generality 
explicit is, after all, precisely the function of a general theory: 
assuming, that is, that such generality really exists. 

Secondly, activities in formal organizations are clearly "mo- 
tivated," i.e., oriented toward the achievement of some goal. The 
problem arises of the extent to which, and through what means, 
the goals as well as the activities of units at various levels need to 
be integrated. While Parsons regards goal orientation as the dis- 
tinguishing feature of all social action, common to units at all 
levels (SS 4-5), he defines organizations as systems which give 
primacy to goal attainment (ASQ 64). Once again, such goal 
orientation exists not only for the individual roles constituting an 
organization ( at least, this is true of formal role prescriptions ) , but 
also— though we get onto more controversial ground here— it is true 
of the organization taken as a unit, and of the system, such as the 
educational system or the economy to which the organization 
belongs (though here it is more appropriate to speak of the per- 
formance of a function rather than to reify and speak of seeking 
a goal).* 

Finally, organizations, more than other kinds of aggregates, have 
explicit mechanisms for solving the twin problems of how to main- 
tain their identity vis-a-vis their environment, maintaining whatever 
patterns of internal relationships they have established, while at 
the same time obtaining from the environment the support they 
need for survival. 

According to Parsons, the central task of the social sciences is 

* A. W. Gouldner has drawn attention to the fact that this distinction, 
which is very much of a difference, is shghted in this part of Parsons* 
work. See his "Organizational Analysis," in Robert K. Merton, Leonard 
Broom and Leonard S. Cottrell, Jr., eds., Sociology Today: Problems 
and Prospects (New York: Basic Books, 1959), Chapter 18, pp. 

216 • Henry A. Landsherger 

the formulation of a single theory applicable, after appropriate 
specification, to all kinds of social systems. Such a theory would 
show (a) how individually motivated units of such systems can 
attain their private ends while (b) simultaneously furthering the 
collective (i.e., the system's) end, (c) maintaining stable relation- 
ships with other units, and (d) remaining integrated both within 
themselves and with higher and lower level units. As has been 
pointed out, formal organizations face these problems. Since 
they have been subjected to study, one would expect Parsons to 
look to students of organization to provide him with data, and with 
theories, so readily transmutable into his conceptual scheme as 
to make it apparent that his framework does, indeed, fit the exist- 
ing state of knowledge in this field even though it was not specifi- 
cally designed to do so. This, after all, is the true test of a general 
theory, and Parsons might be expected to use existing research 
findings to prove his claim of generality. 

Parsons has not done the expected. He has admittedly attempted 
to apply his general scheme to organizations. But he has done so 
neither very extensively, nor very systematically. Nor has he based 
his writings on an explicit examination of the work of others in 
an attempt to show point-by-point congruence with his own ideas. 

As for extent. Parsons dealt thoroughly with organizations for 
the first time in 1956, in two relatively short articles in the Adminis- 
trative Science Quarterly (ASQ).* He has followed these with two 
further contributions, one of which in particular is a substantial 
elaboration of his earlier statement, f 

As for being systematic, Parsons discusses extensively in his 
other works, yet mentions only briefly in his two main articles, three 

* Talcott Parsons, "Suggestions for a Sociological Approach to the 
Theory of Organizations," Administrative Science Quarterly, Vol. 1, 
Nos. 1 and 2, June, September 1956, pp. 63-85 and 224-39. 
f Talcott Parsons, "The Mental Hospital as a Type of Organization," 
in Milton Greenblatt, Daniel Levinson, and Richard H. Williams, eds., 
The Patient and the Mental Hospital (Glencoe, 111.: Free Press, 1957), 
Chapter 7, pp. 109-29; and "Some Ingredients of a Generar Theory 
of Formal Organization," in Andrew W. Halpin, ed., pp. 40-72 in 
Administrative Theory in Education (Midwest Administrative Center, 
University of Chicago, 1958), Chapter III, pp. 40-72. 

parsons' theory of organizations • 217 

concepts which he regards as intimately related to the concept of 
organization. These are the concepts of the occupational system 
(WP 254-64; as well as scattered references, some of considerable 
length, in FSI, SS, and ES); the contract; and the labor market 
(for the latter two, see particularly ES 114-122; 144 et seq.; 175 
et seq.). In addition, all his concepts, being part of a general theory, 
should by definition be relevant to organizations. 

Finally, Parsons' writings, in this as in other fields, do not take 
the form of an explicit comparison of his formulations with existing 
ones. Parsons seems well acquainted with at least some of the key 
problems and controversies in organization theory, though there are 
indications that his knowledge of the field is neither exhaustive nor 
systematic. He has certainly put forward some provocative ideas 
about the few problems which he has discussed in detail. 

The Relevance of Parsons Work to Organization Theory 

The student of organization should in his turn look eagerly to 
Parsons. He should be interested to see whether Parsons' general 
theory contains solutions to theoretical (or, for that matter, prac- 
tical ) problems which have not been solved in the field of organiza- 
tion theory, but which have been solved, mutatis mutandis, in 
other special fields. For example: does Parsons' theory of the divi- 
sion of leadership into "external task" and "internal integrative" 
functions, based on his studies of the family and of small groups, 
really tell us anything we did not know about leadership in complex 
organizations? Do his theories at least correspond to existing knowl- 
edge in the field, which would be achievement enough? 

If Parsons on his side has not gone out of his way to forge strong 
links with organization theorists, have the latter at least tried 
to meet him halfway? Surprising though it may seem, since Parsons 
is, after all, not primarily an organization theorist, his influence has 
been noticeable and is growing. His ideas have been used most by 
the youngest areas of research: research into hospital and educa- 
tional organizations— perhaps because no established theoretical posi- 
tions existed here. Thus, he is the author most frequently cited in 


218 • Henry A. Landsherger 

the index to a collection of studies on the role of the school super- 
intendent.* Similarly, the summary essay in a recent volume of 
readings dealing with the patient-mental hospital relationship con- 
cludes that Parsons' theory of action mediates best between the 
wide range of concepts from sociological to psychoanalytic theory 
which it is necessary to employ in order to understand this field. f 
Parsons is not, however, referred to favorably in any other contribu- 
tion to this volume except his own. 

Studies in the more established areas of research— into industrial 
and governmental organization— have made less use of his writings. 
There is, moreover, no major study devoted wholly to a test of his 
theory, in the manner in which entire studies have been designed 
to substantiate or refute Max Weber. Typically, those writers who 
have used Parsons at all will make references to only a limited 
number of his concepts. His definitions of role and role conflict, 
and his pattern variables— especially the universalism— particular- 
ism distinction— have been the most popular. J This kind of selective 
use of his concepts is growing even in industrial research. The find- 
ings of a recent, very specialized study in psycholinguistics, appear- 
ing in as surprising a home for general sociological theory as the 
Journal of Applied Psychology, were interpreted as confirming in 
an industrial setting that dual function theory of leadership to 
which I referred earlier. § 

However unsystematic and scattered they may be, therefore, the 
links between Parsons' work and the work of others in the field of 
organization theory are growing stronger. It is the purpose of this 
paper to examine Parsons' work in order to evaluate, more compre- 
hensively, the nature of his potential contribution to the study of 

* Neal Gross, Ward S. Mason and Alexander W. McEachum, Explora- 
tions in Role Analysis (New York: John Wiley and Sons, 1958). 
f Richard H. Williams, "Implications for Theory," in Greenblatt 
et al., op. cit.. Chapter 37, pp. 620-32. 

I See, for example, Peter M. Blau, Dynamics of Bureaucracy 
(Chicago: University of Chicago Press, 1955). 

§ Harry C. Triandis, "Categories of Thought of Managers, Clerks, 
and Workers about Jobs and People in an Industry," Journal of Ap- 
plied Psychology, 43:338-43 (October 1959). 

parsons' theory of organizations • 219 


System Problems and Pattern Variables 

To understand Parsons' definition and description of organiza- 
tions, and the special place they occupy in his general analysis of 
social systems, it is necessary to review briefly his two sets of 
basic concepts— the four system problems and the five bi-polar 
pairs of pattern variables. A more thorough treatment will be found 
in the earlier papers in this volume, particularly those by Devereux 
and by Morse. 

The system problems are, it will be recalled, those of Adaptation, 
Goal Attainment, Integration, and Latency, the latter also referred 
to as Tension Management and Pattern Maintenance. Maximizing 
efforts to resolve one problem, according to Parsons, intensifies one 
of the other problems. Resolving adaptation problems increases 
problems of integration (and vice versa); goal attainment inten- 
sifies the problem of general pattern maintenance; and there is a 
more general antagonism between the resolution of A and G (task 
or instrumental) problems on the one hand, and the I and L prob- 
lems (social-emotional or expressive) on the other. In a decision- 
making group, for example, when its members cooperate in the 
task problems of gathering and digesting information (A) prior 
to making a decision (the group's goal— G), they will strain their 
personal relations to each other (I), and temporarily prevent indi- 
vidual members from fulfilling other needs and obligations (L 
problem ) . 

The pattern variables which describe the different kinds of rela- 
tionships during each of these problem-solving phases mirror the 
dilemmas in the phases themselves. The adaptive phase requires 
that persons do not attempt to use their relationship for gratifica- 
tion but to further the distant goal (neutrality); that they treat 
others on the basis of what they can contribute (performance); 
that they restrict their relationship to others to the narrow front 
demanded by matters of "business" (specificity); and that they 
ignore the particular relation in which a person stands to them 
outside the "task" setting ( universalism ) . Social integration, on the 


220 • Henry A. Landsberger 

other hand, requires relationships to be structured in the opposite 
way, emphasizing a different value system : aff ectivity— by definition, 
the relationship during integration should be enjoyed per se; 
quality— the group needs to value its members simply because they 
are members, not only for what they can do; diffuseness— during 
integration, a wider range of the personality becomes involved than 
during "business," when only limited skills are drawn upon; and 
particularism— persons are valued because they are members of 
ego's group. 

The fifth pattern variable (collectivity vs. self -orientation ) is 
qualitatively different from the other four. It represents a scale for 
measuring the extent to which the unit (the role in a small group; 
the small group in the larger aggregate to which it belongs) acts 
or should act on behalf of the superordinate system. If the uni- 
versity professor is expected to be guided explicitly, by what is 
good for his department as well as by what is helpful to his role 
specifically, the role of professor is collectivity oriented. If the 
department chairman is expected to fight for his department only, 
and let the interests of the university as a whole be defended by 
dean, provost or president, the role of departmental chairman is 
self oriented— the "self," at this level, being the department, not 
the role and certainly not the individual. Once it has been estab- 
lished on behalf of which system (i.e., at what level, its own or the 
next higher one) a unit is really acting, its relationship to its role 
partner can be analyzed in terms of the other four pattern variables. 

Certain characteristics of these concepts, and of the way in 
which Parsons uses them, should be briefly noted. First, he sees the 
four problems not only as facing all systems, but also as being dimen- 
sions which can be used to describe the relationships ("boundary 
processes") between systems and between sub-systems of a system. 
As we shall see, the occupational contract through which organiza- 
tions obtain most of their human resources, has components that 
are adaptive, integrative, and so on, each of which really covers a 
separate subcontract. Moreover, there is some tendency for the 
solution of the four problems to be attempted in a temporal 
sequence. At one time the solution of one problem is emphasized 
more than it is at all other times, then the next one, and so forth, 
until the first problem recurs. Hence the system problems and the 

parsons' theory of organizations • 221 

specific pattern variables associated with them are characteristic of 
temporal stages. Further, to deal with these problems, Parsons be- 
lieves that systems develop sub-structures each of which specializes 
in the solution of one of them ( but which in turn has four sectors ) . It 
is particularly important for an understanding of Parsons' treat- 
ment of organizations to understand that he sees one system as 
having sub-systems, and as itself being the sub-system of the next 
higher system, greatly influenced by its functional contribution to 
the higher system, and the function of the higher system itself. 

The system problems exert, therefore, not only a temporal influ- 
ence, but also a major influence on the structural differentiations 
of a system at any given point in time, and on the flows across its 
boundaries to other systems: higher, lower or at the same level. 

The pattern variables are hkewise used not only to describe and 
categorize existing relationships, but also to describe the norms 
governing relationships and hence the types of deviance which 
may occur. They describe four basic motives, and hence the incen- 
tives and sanctions which can be used to control personalities. 

The reader's sense of having placed before him a completely 
integrated theoretical system to cover all levels of society and all 
its activities is, of course, greatly enhanced by the use of this 
standard terminology. While details of definition and of termi- 
nological usage are changed (often without adequate notice to 
the reader) Parsons' underlying ideas have remained constant. 
Parsons would, of course, claim that the theoretical integration he 
has achieved is based on the discovery and elucidation of genuine 
underlying uniformity of process and problem, and not on a mere 
terminological Gleichschaltung. 

Organization and Collectivity Orientation 

Parsons uses both the pattern variables and the four system 
problems to define organizations. The fit is at times not very good 
and one wonders whether, contrary to the nature of the material 
with which he is dealing. Parsons was forced or lured into the use 
of these concepts in order to maintain his claim that they are 
suitable for all purposes. However, there is often also a hisjhlv 
suggestive kernel hidden in what appears otherwise as a somewhat 

222 • Henry A. Landsberger 

forced set of ideas. This is well illustrated by the use he makes of 
one of the pattern variables— self- vs. collectivity-orientation— in 
defining organizations. 

In his earlier writings (i.e., in the Social System and in Toward a 
General Theory of Action) Parsons defined organizations as one 
kind of "collectivity ready to act in concert to achieve a goal," 
"Gemeinschaft" being the other kind. By this definition, an organi- 
zation had an exceptionally high degree of collectivity orientation 
and "solidarity." The lowest degree (apart from undefined role 
"intermeshing" ) was "role integration" where— as in the case of 
buyers and sellers in a free market— each role partner merely knows 
what the behavior of the other means. The next higher level is a 
"collectivity," defined as a role system in which members regard 
certain actions as required in the interest of the integrity and 
continuity of the system, engaging in these activities regardless of 
the immediate self-interests of the role. This, of course, fits with 
the idea of collectivity orientation on the part of a role— the role 
takes on responsibility for the continuity of the system. Next come 
collectivities ready to act in concert to achieve a goal which is 
shared (i.e., widely held) and collective (i.e., "gratifying to mem- 
bers other than, but including, the actor," GTA 192). An organi- 
zation is one variant of such a collectivity -ready-to-act-in-concert: 
where the goal is the result of the collectivity's action, not inherent 
in the action, as is the case with the other variant, a "Gemeinschaft." 
A group of amateur folk dancers are a ''Gemeinschaft" because 
their goal is the activity of dancing itself. Their most important 
activities are of the L and I kind, i.e., they are biased in the 
''expressive" direction. On the other hand, a group dedicated to 
nuclear disarmament would be an organization, since its most 
important activities are of the A and G variety (it is biased in the 
"instrumental" direction) and its goal is merely the result of these 
activities, not inherent in them. Schools, too, are organizations, 
since their goal is the socialized individual, not the educational 
process per se; and collectivities in the economic (i.e., adaptive) 
sector of society are par excellence likely to be "organizations" by 
virtue of their position in a sector in which the goal is the produc- 
tion of facilities for goal achievement elsewhere. 

Parsons describes both organizations and Gemeinschaften as 

parsons' theory of organizations • 223 

emphasizing goals in the eyes of their member-units. But he 
regards only organizations as being biased in the goal attainment 
(or, more broadly, instrumental) direction. This paradox is resolved 
through taking account of Parsons' system-within-system-within- 
system manner of thinking. The definition of the concept "goal 
attainment," and hence of organization, is basically not descriptive 
of the motivation of the lower units, but is in terms of the system 
superordinate to the organization. Parsons' definition highlights the 
point that, in a purely functional analysis of society, it is irrelevant 
that members of, e.g., a group devoted to nuclear disarmament 
might enjoy being together (I), or even that its members may get 
satisfaction from instrumental activities such as obtaining signa- 
tures. The significance to society is the effect which the group's 
ultimate product— pressure— has on other systems. To be classed 
as au organization, a group merely needs to act "as if" it were 
goal oriented, by actually affecting another system. 

While such functional analysis is, in itself, acceptable and useful, 
it is insuflScient as an explanation of such well concerted activities. 
It poses an obvious next question: for whom is the organization's 
goal a motivational goal? The answer to this question has ideo- 
logical overtones. On one side are the social critics, who believe 
that many organizations serve the ends of the small elite who 
control these organizations and possibly society at large. Critics 
feel that the very phrase "the goals of the organization," or any 
reference to the social utility of an organization's product, glosses 
over the fact of self-serving by a power elite. On the other side 
are those who believe that cooperation for a common goal is, or 
at least could be, more than a slogan for all members of an 
organization. The rise of labor unions in the economic sector, sym- 
bolizing that goals are not fully shared throughout these organiza- 
tions at least, has made the answer to the question. Whose goals? 
even more intriguing. 

Parsons' position on this critical issue is ambiguous, despite the 
fact that he is well aware of the inadequacies of functional explana- 
tions. At times he seems to be sidestepping the controversy about 
individual motivation by asserting that he is not describing what 
exists, but describing logical possibilities and logically pure cases 
within his set of concepts, or that he is describing roles, not flesh 

224 • Henry A. Landsberger 

and blood individuals, in a smoothly functioning model. Logically, 
certainly, there is the possibility of a collectivity oriented set of 
roles engaged in instrumental activity in pursuit of a goal which is 
"shared" and "collective." 

Parsons' later formulations make it clear, however, that he uses 
intense collectivity orientation as more than just a defining char- 
acteristic of a logically possible entity. He uses it as an explanatory 
concept and to describe what actually exists in the eyes of members 
of the organization. Thus, Parsons explains that centralized deci- 
sion-making by leaders of organizations is legitimized in the eyes 
of followers: 

... by the expectation that management will be competent and 
that there will be an identity of interest between management 
and other employees in giving management the power it needs 
to do the job effectively subject to fair treatment of employees. 
(ASQ 234-5) 

This astonishing description of the employee's state of mind is 
made more credible by Parsons' explanation that the coincidence 
is neither spontaneous nor due to managerial good will, but 

... is controlled externally, first by competition with other 
firms, so that presumably an ineffective management would not 
be able to continue in business, and secondly by the free labor 
market to the extent that its employees are free to quit and 
seek other employment. (ASQ 235) 

This statement, while more acceptable, hardly portrays a situation 
marked by genuine collectivity orientation, thereby raising a ques- 
tion as to how fuzzy the concept still is. The above quotation means, 
in effect, that workers, through the use of "self-oriented" market 
mechanisms such as quitting, can limit the extent to which manage- 
ment dare go against workers' interests. Under these protective 
conditions, they are willing to take specific management decisions 
on trust. Clearly, this power of control disappears in a market where 
labor is surplus, and the coincidence of interests between manage- 
ment and labor will shrink correspondingly, leaving the existence 
of organizations once more unexplainable in terms of collectivity 

parsons' theory of organizations • 225 

Parsons also uses the concept in what appears to be a third, in- 
between sense, postulating collectivity orientation, or a minimal 
amount of it, as a prerequisite if certain groups are to be in equilib- 
rium and survive. In this mood, discussing equilibrium and stabil- 
ity in general, Parsons will state that the concept of equilibrium is 
not to be used as "an empirical generalization." (SS 481) He also 
makes an important distinction between role prescriptions and the 
individual's actual motivation. He asserts that for a collectivity 
merely to survive, certain roles must value collectivity orientation 
and there must be a degree of coincidence "of a set of common 
value patterns with the internalized need-disposition structure of 
the constituent personalities." (SS 42) Talking specifically about 
organizations, he refers repeatedly to the likelihood of friction 
between personality on the one hand and role requirements on 
the other, (e.g., ES 177) He deems it unlikely that organizations 
can set up mechanisms which will successfully counter (let alone 
prevent) "an inherent centrifugal tendency . . . deriving from the 
personalities of the participants, from the social adaptive exigencies 
of their particular job situations; and possibly from other sources." 
(ASQ 79) It is in this more realistic mood, too, that he appreciates 
that it is psychologically very difficult for employees to comply with 
the edicts of far removed leaders. (SS 279-80) This motivational 
vacuum is filled by the growth of informal role expectations, leading 
to role conflict with formal role expectations. This insight and its 
conceptualization is in line with some of the most recent research 
on the subject of informal organization.* It is, unfortunately for 
Parsons, only loosely connected to his main conceptual scheme. 

The contradictions between these statements stressing the unlike- 
lihood of deeply felt collectivity orientation and previous statements 
asserting the coincidence of management and employee interests, 
exemplify a general tendency by Parsons to glide, imperceptibly, 
from the description of a possible model and a definition of its 
various parts to statements concerning conditions and relationships 
necessary and existing if a certain system is to be stable and then 
to assertions about phenomena and their relationships as they 

* Melville Dalton, Men Who Manage (New York: John Wiley and 
Sons, 1959). 

226 • Henry A. Landsberger 

actually exist. Parsons' statements become more suggestive and 
more acceptable if the reader is clear within which of these three 
possible levels of discourse Parsons is moving at any one time. 
Most of Parsons' statements seem to fall in the middle category, 
postulating conditions which need to exist to a greater or lesser 
extent if a social system is to continue in being. 

The idea of collectivity orientation, for example, makes more 
sense and becomes an exciting idea for research if it is not taken 
as describing relationships between members in all existing organi- 
zations. Instead, it should be regarded as conditional, and hence 
quantitatively related to the greater or lesser success of organiza- 
tions. Or one might think of a minimal amount of collectivity orien- 
tation as required for their survival. Fouriezos, Hutt, and Guetzkow,* 
observing decision-making groups in government and industry, have 
indeed found a negative correlation between group productivity 
and average level of self-orientation. They are impressed with the 
importance of self- and collectivity-orientation as a causal variable 
of goal attainment, and plead that more research be done on this 
concept, difficult though it is in practice to operationalize it. 

Parsons believes that managers in particular need to be col- 
lectivity oriented, both because they are responsible for the organi- 
zation as a whole, and because employees' perception of the extent 
of management's collectivity orientation affects their readiness to 
obey. Ironically, one of Parsons' severest critics, C. Wright Mills, 
quotes Werner Sombart, Walter Rathenau, and others to support 
his contention that managers are in fact collectivity oriented to a 
degree that makes it seem as if organizations had motives of their 
own.f But there is only one empirical study, Dalton's, linking the 
degree of management's collectivity orientation to organizational 
success (Dalton maintains that there is a negative correlation!). 
No studies have linked the degree of management's collectivity 
orientation to subordinates' obedience: a hypothesis which makes 

* Nicholas T. Fouriezos, Max L. Hutt, and Harold Guetzkow, "Self- 
Oriented Needs in Discusson Groups," in Dorwin Cartwright and 
Alvin Zander, eds.. Group Dynamics: Research and Theory (Evanston, 
III: Row, Peterson and Co., 1953), Ghapter 24, pp. 354-60. 
f G. Wright Mills, White Collar: The American Middle Classes ( New 
York: Oxford University Press, Galaxy Books, 1956), pp. 107-8. 

parsons' theory of organizations • 227 

a great deal of sense for organizations not in the economic sector 
(as Parsons realizes). No studies exist on whether, as Parsons 
believes, collectivity orientation is less important for some roles 
than for others: e.g., roles which represent the organization in its 
dealings with the environment. Is collectivity orientation more im- 
portant in some organizations and in some situations than in others? 
Are organizations less successful in cultures which do not value 
collectivity orientation in organizational settings? What makes 
organizations, as units, more or less collectivity oriented, taking 
responsibility for the solution of problems facing the system as a 
whole? Concretely— under what circumstances will universities re- 
frain from competing for graduate students, since it serves their 
own ends without adding to the product of the educational system 
as a whole? All these questions would profit from investigation. 
Parsons' pattern variable has the merit of raising them and of 
drawing attention to the possibility of underlying similarities, de- 
spite the tremendous difference in levels of social analysis and in 
the contexts of different organizations. 

One final point concerning those instances in which Parsons is at 
the level of describing what exists. The precarious balance which 
he sees between social requirements and actual motivation, and 
between the requirements of existing systems and those of emerg- 
ing systems, are applied not only to organizations, but also to society 
at large. In this more realistic mood, Parsons seems fully aware of the 
role which existing elites may play in causing revolutions through 
an attempt to preserve an existing system intact. (SS 520) The 
fact that he deliberately disassociates himself from a "predominant 
factor theory" (i.e., economic theory) of social change (GTA 232-3) 
should not mislead one into accusing Parsons of being oblivious to 
the existence of powerful forces and groups within society which 
continually modify it in a more or less drastic manner. 

The Structure of Organizations and the Four System Problems 

Parsons' most substantial foray into organization theory is an 
attempt to reformulate it in terms of his four system problems. 

The latency problem (L) concerns, as we would expect, the 
integration of the organization with higher order patterns of cul- 

228 • Henry A. Landsberger 

ture values. This is achieved, most importantly, through interpret- 
ing the particular goal of the organization ( for example, production 
of roller bearings, or education in modern languages) as being 
congruent with some value of the next larger social system. Culture 
for a lower system consists, then, among other ingredients, of the 
values of the next higher system. This constitutes "the legitimation 
of [the organization's] place or 'role' in the superordinate system" 
(ASQ 68). Such legitimation helps in the internal running of the 
organization. It also enables the organization, when faced with 
outside pressures, to assert the primacy of its goal over other 
possible goals. For example, an educational organization can assert 
the primacy of producing educated citizens over creating good will 
in the community (I) and over the production of facilities and the 
making of a profit (A). Functions cannot be accomplished (it will 
be more difficult to "produce" the goal) unless culture values 
recognize the goal to be a legitimate one. 

If one thinks of economic organizations only, the importance of 
such legitimation is not great, since economic organizations obtain 
most of their resources through funds obtained from the sale of 
their product. But failure to have an accepted goal can break an 
organization which is not in the economic sector. Mental health 
organizations, shunned in many communities, are short both of 
financial support and of the more diffuse nonfinancial support ( e.g., 
volunteers, members) which is just as important to them. Heart 
associations have fewer troubles. The success with which L prob- 
lems are solved therefore affects the solution of other organization 
problems— obtaining resources ( A ) , designing an effective ( G ) and 
cohesive (I) internal structure. This is reflected in the fact that 
each of these three sectors in turn have their own L sub-sectors, as 
we shall see. 

The adaptive problems of the organization (A) concern its ef- 
forts at the "mobilization of resources," the acquisition of resources 
from the environment. These resources are the traditional four 
factors of production: capital (Aa), labor (Ag), entrepreneurship- 
organizing ability (Ai), and "land" (Al). 

These resources are chiefly obtained through contracts. Since 
contracts are indicative of how, in general. Parsons handles flows 
between systems, we shall examine one of them— that for labor— 

parsons' theory of organizations • 229 

at length in the final section of this paper. Here we will briefly 
look at the one resource— Al, input on value grounds— which by its 
nature is not itself a resource contracted for, and operates through 
its effect on the ease or difficulty with which contracts for other 
resources can be made. As expected, Parsons links the Al sub-sector 
with the organization's larger L (value) sector, and thereby again 
with larger social values. The Al resources are therefore those which 
are committed to the organization on basic value grounds; they do 
not have to be rewarded in the short run in order to be retained. 

Economists have for long recognized that certain resources are 
"committed" to a sphere of production in the sense of their being 
immobile and unable to gain a comparable income elsewhere. Good 
land has typically been regarded as of this kind. High rent is paid 
for it because it is productive and hence farmers compete for it— 
but the leasor could be forced to accept a far lower rent because 
there may be nothing else for which he can use his land. Economists 
have recognized that resources other than land— e.g., labor— may be 
similarly immobile, and may therefore be earning a "quasi-rent." 

It seems that Parsons has broadened and changed this concept 
even further, first by including also services which are not rewarded 
by the payment of quasi-rent, and by looking at social values, not 
only immobility, as the cause of commitment. For convenience, 
economists confine their considerations to exchanges and flows in 
which money is involved— hence the domestic work of wives is 
excluded from calculations of the gross national product. From a 
social point of view, a productive contribution remains such whether 
or not it is financially rewarded. Parsons refuses to be bound by 
an arbitrary dividing line— with some attendant disadvantages, of 

Parsons has also broadened the concept of commitment by 
seeing it as operative at all levels: society as a whole, the level 
of social sub-sectors, and that of the organization. Thus, at the 
level of the U. S. economy as a whole, labor may have a higher 
motivational commitment to work on value grounds ( the phenome- 
non of the Protestant ethic ) than is the case in other societies which 
do not have this skewing of values in the adaptive direction. On 
the other hand, the U. S. may have at its disposal less motivational 
commitment to teach than other societies (partly because of lesser 

230 • Henry A. Landsberger 

basic commitment to L values ) . These are interesting ways to think 
about frequently mentioned diflferences in cultural values, and of 
integrating them conceptually with other factors of production. 
There is a counterpart of this value commitment even at the level 
at which we are talking— that of the organization. It, too, may have 
at its disposal labor loyalties (faithful employees), entrepreneurs, 
and capital, for which it need not pay in the short run. 

The Goal-attainment sector (G) of the organization is called by 
Parsons "The Mechanism of Implementation." It will be remem- 
bered that the G sector of society is the power sector. So also is 
the G sector of the organization— the power to mobilize resources 
for goal attainment. The Gg sector concerns major "Policy Deci- 
sions" about how the goal is to be attained— the nature and quality 
of the product, change in scale of operations, and "organization- 
wide problems of modes of internal operation." This type of deci- 
sion is a very serious one and commits the organization. Selznick 
has called the making of this kind of decision "leadership," point- 
ing out that over and beyond routine administration, decisions about 
the character of the organization and the internal distribution of 
power need to be made if the organization's purpose is to be ful- 

The Ga sector ("Allocative Decisions") concerns lower level 
decisions of allocating responsibilities and financial resources among 
personnel. This is Selznick's "administration." Parsons' scheme makes 
automatic provisions for a distinction which is fruitful, but which 
very few writers have made. 

The Gi sector ("Coordinating Decisions") deal particularly with 
personnel who, unlike financial resources, have to be motivated and 
hence make the coordination of activities difficult. The means for 
accomplishing coordination are penalties (coercion), rewards (in- 
ducement), and "therapy." The latter is defined as operating 
directly to change the motivations of individuals rather than taking 
them for granted and offering either rewards or penalties. 

The Gl sector again involves values: the values covering, legiti- 

* Philip Selznick, Leadership in Administration ( Evanston, 111. : Row, 
Peterson and Co., 1957). 

parsons' theory of organizations • 231 

mating, "authorizing" the measures and decisions involved in the 
other three sub-areas of G. 

By definition, the organization is a social system emphasizing its 
own G sector, signifying that the nature of its goal greatly affects 
its other sectors— more so, presumably, than they do it. 

The I sector, as Parsons discusses it in the Administrative Science 
'Quaiiierly (p. 80 S), concerns "the mechanisms by which the or- 
ganization is integrated with other organizations and other types of 
collectivity in the total social system." Here, Parsons clearly acci- 
dentally moved up one of his own levels: "I" should refer to the 
integration of the system within itself, not to the integration of the 
system with other systems (which is the I problem for the next 
higher level). Parsons does deal with internal integration, but, as 
we have seen, under the heading of Gi. 

The problems he rapidly lists under the "I" heading are, however, 
interesting ones. In the la sector are worked out the problems that 
arise because resources have to fulfill their other role obligations. 
For example, only if employees are limited to, let us say, eight 
liours of work for the organization can they— and, therefore, ulti- 
mately, the organization— remain integrated with the family. The 
Ig sector deals with limitations on power and authority so as to 
retain congruence with outside obligations; and li deals with the 
integration of the organization's integrative rules— i.e., the fact 
that there are certain rules which are general throughout society. 
They therefore integrate the organization with the rest of society. 
An example is the general prohibition against the use of force 
against the person. Neither contract nor authorization can abrogate 
these general rules. 

An evaluation of this scheme would have to list two deficiencies, 
and one major point of attraction. On the negative side, this 
attempted application of the four system problems brings out force- 
fully how ambiguously one concept is tied to another and to its 
empirical referent. It is not in all cases clear to me how the con- 
cepts "boundaries," "flows," "sectors," and "sub-sectors" are related 
to each other, nor how functional sectors of society are related to 
the concrete organizations through which the functions are ful- 
filled. Do all the resources needed by a system come in through 
the A sector and its boundaries even though they end up elsewhere 

232 • Henry A. Landsberger 

in the system? It would seem so from Parsons' analysis of organiza- 
tions. Yet in Economy and Society, the I sector of the economy 
obtains some of its resources directly from the I sector of society 
(ES 68, Fig. 4). 

As for the actual four system problems, I am not convinced that 
some of them may not need to be merged and others added. Parsons 
has on occasion linked A and G together, avoiding the need to make 
a distinction between them. Since organizations are admittedly goal- 
oriented, but since one tends to think of the ideal type as being 
governed by A values (neutrality, universalism, specificity and per- 
formance), merging the two sectors would remove the need for 
deciding whether organizations are A or G biased. In any case, 
there seems to be some lack of clarity here. 

The case of the entrepreneur may be cited as another example of 
lack of precision in the empirical referent of a concept. In Economy 
and Society Parsons saw the entrepreneur as contributing integra- 
tive ability to the economy. Yet in Parsons' discussion of organiza- 
tions, the kinds of activities in which, certainly, executives engage 
—making policy decisions, allocating budgets and responsibilities- 
are placed in the G sector, which one would have expected to be, 
devoted to problems of financing, since Parsons regards finance as 
an important form of power, and power supposedly belongs in the 
G sector. Nor does the organization's integrative sector, as we saw^ 
Parsons describe it or as conceived in any other way, seem to be 
the field for the kinds of integration of capital and labor which 
Parsons originally had in mind when talking about entrepreneurs. 

The reader should be warned, however, that even if unnecessary 
vagueness were removed, many of these concepts would still not 
correspond to the kinds of subdivisions of organization with which 
he is familiar— sales, production, accounting, and the like. There 
obviously is no concrete latency sector nor a concrete latency activ- 
ity or flow validating an organization's goal or its operative roles 
against the larger goals and rules of society. The personnel depart- 
ment does have substantial responsibility for acquiring labor re- 
sources, but it also does more, while on the other hand the commit- 
ment of labor on value grounds is beyond its scope. For the most 
part, Parsons' concepts are analytical features which can be used 
in a highly partial description and interpretation of a variety of 

parsons' theory of organizations • 233 

actual occurrences. In itself, this analytical abstractive character 
is not at all unscientific— quite the contrary. It is unusual and strange 
in a field in which much of the writing is in the form of descrip- 
tions of total events with relatively little abstraction. 

The second criticism of the scheme, apart from its vagueness, is 
that it does not as yet constitute a set of hypotheses. Parsons him- 
self is at one point fully aware of this: 

It is extremely important to be clear that what we have pre- 
sented ... is a paradigm and not a theory, in the usual sense 
of the latter tenn as a system of laws. . . . We have had to 
formulate the concepts of motivational process as mechanisms, 
not as laws. (SS 485) 

Parsons says this is due to paucity of knowledge of laws— but adds 
that such knowledge was just enough to formulate the mechanisms 
(e.g., that hierarchical distance between persons aflFects their motive 
to obey each other). The systematic organization of the "mecha- 
nisms" of which his work consists is, therefore, intended to sum- 
marize knowledge to the extent that we have it, and to "give us 
canons for the significant statement of problems for research so that 
knowledge of laws can be extended." (SS 485) 

The various sectors and sub-sectors are primarily systematically 
arranged pigeonholes for the ordering of problems. An organiza- 
tion's product may be analyzed from the point of view of its 
relation to society's goals and values. One may even speak of the 
organization's having a problem legitimating its product in terms of 
larger social values. But these are not hypotheses. They lead, at 
most, to statements about minimal conditions necessary for survival, 
with no actual minima quantitatively specified. As in the case of 
the need for collectivity orientation in organizations, where Parsons 
at times believed that the minimally needed quantity was high, at 
other times that it was low, similarly in other instances the reader 
is not sure to what extent the various problems really do need to 
be solved. He is not given predictive hypotheses covering how and 
under what conditions they will be solved. Parsons makes no 
attempt to apply to organizations one of the few hypotheses built 
into his scheme— the phase hypothesis. This would lead one to 
predict certain changes in organizations over time. Very few 


234 • Henry A. Landsberger 

writers mention the possibility of such systematic changes occurring 
—and they are, Hke Selznick, among the very best in the field. 
Parsons' unstated hypothesis is not a bad one, that organizations 
will veer between efficiency (A and G) on the one hand, and the 
stability and possible drag on efficiency signified by having a well 
integrated organization with articulated values (I and L) on the 

Against the conceptual vagueness and the difficulty of deducing 
hypotheses from Parsons' scheme need to be put both its compre- 
hensive nature and the fact that hypotheses can easily be stated 
in terms of it and are at least suggested by it if not deducible 
from it. 

About its comprehensiveness there can be no question, nor about 
the pedagogical value of such comprehensiveness for a field which 
should never have become as fragmented as it is. No other field 
in the behavioral sciences should cut more freely across disciplines 
—particularly in order to integrate economics— than that of organi- 
zation theory. No other field in the behavioral sciences needs to be 
as urgently in contact both with the societal (macro-) level and 
with that of the small group and the individual. Parsons' problem 
areas, if they do nothing else, imply an assertion that organization 
theory is of this kind and that levels and disciplines cannot simply 
be given separate boxes as "additional factors," but need to be 
seen as linked to each other, or as special cases of each other, or 
positively fused together, as in the case of the labor contract and 
the effects of social values on it. In three areas in particular, Parsons' 
comprehensiveness brings together approaches to organization 
which should never have been separated. 

First, unlike the writings of many other sociologists. Parsons does 
not neglect traditional administrative "theory" as formulated by 
Urwick and his followers.* This approach emphasizes that an organ- 
ization could not be run without such processes as policy making, 
organizing to execute the policy, coordinating and controlling 
activities, delegating tasks and authority, making provisions for 

* See, for example, "Notes on the Theory of Organization," in L. 
Gulick and L. Urwick, eds., Papers on the Science of Administration 
(New York Institute of Pubhc Administration, Columbia Univ., 1937). 

parsons' theory of organizations • 235 

communications, and so on. Parsons covers this under the heading 
"Ga." The term "adaptation" may seem a httle forced. But pre- 
sumably—with some merit— it refers to the fact that the individual's 
potential for relating himself to others is being adapted to the 
organization's purpose by being controlled and formalized, guided 
as to when and when not to communicate, and so on. 

Sociologists have for the most part ignored the formal aspects of 
organization as objects of study for their own sake, as Gouldner 
has recently pointed out.* This is surprising, since much sociological 
work on organizations was inspired by Weber, who accented 
heavily the bureaucracy's need for precise definition and limitation 
of duties, the allocation of these duties among "offices" (roles), 
the hierarchical arrangement of roles and their subjection to imper- 
sonal and limited authority, and so on.f 

From this starting point, however, the predominant movement 
of sociologists has been in an essentially psychological direction, 
focusing on the interpersonal and intrapsychic eflPects of organiza- 
tional structure, on how informal organization has helped or 
hindered formal organization, not on formal organization itself. 
Investigations have dealt with the reasons for alliances and feuds 
(neither sanctioned by the formal organization structure) between 
engineers and line officials in industrial plants; J or with the 
growth of a more formalized communication system out of the 
temporary anxiety of new managers who do not know the informal 
communication system. § 

Under the heading of the Gi and Gl sub-sectors and the contract 
with which labor resources are acquired. Parsons' scheme allows 
for the study of these manifestations of personality and of informal 
organization. But these headings are a modest few among many 
others, just as in reality they are but one facet of the operation of 

* Gouldner, "Organizational Analysis," in Robert K. Merton, et ah, 

op. cit. 

f See H. H. Gerth and C. Wright Mills, From Max Weber: Essays in 

Sociology (London: Kegan, Paul, Trench and Trubner and Co., Ltd., 

1947), pp. 196ff. 

i Melville Dalton, "Conflict Between Staff and Line Managerial 

Officers," American Sociological Review 15:342-51 (June 1950). 

§ Alvin W. Gouldner, Patterns of Industrial Bureaucracy, Free Press, 

Glencoe, lU., 1954. 

236 • Henry A. Landsberger 

an organization. Parsons accents not only the necessity for policy 
decisions, allocation of budgets, and so on (traditional adminis- 
trative theory, covered by the Gg and Ga sub-sectors) but, in 
particular, he accents the many bonds which the organization has 
with its environment. 

Stressing the influence on the internal structure of organizations 
of (a) the environment within which it operates and (b) the 
nature of its goal and that it has a goal to fulfill are the two other 
merits of the comprehensiveness of Parsons' approach to organiza- 
tion theory. The reader is invariably stimulated to make suggestive 
comparisons between organizations based on differences in total 
cultural environment, the society's sector to which the organiza- 
tions belong, and the nature of their goal. 

The need for an organization's rules to conform in various ways 
to environmental values has been studied all too little by soci- 
ologists, making each of the few existing studies all the more 
precious. Bendix's suggestion that the built-in authoritarianism of 
Weber's image of the ideal type bureaucracy might be the most effi- 
cient way to run a bureaucracy in an authoritarian culture, but not 
in an egalitarian one, is an early speculative venture into this field, 
followed by Richardson's empirical study, contrasting the greater 
egalitarianism of American merchant ships with the greater social 
distance between officers and men on British ships.* The conflict 
between, on the one hand, the stress in Indian culture on the 
family (latency) and hence on quality and particularism even in 
acquiring labor resources for economic organizations, as contrasted 
with an economic organization's need to stress adaptive values 
(universalism and performance) when acquiring resources is graph- 
ically portrayed in a recent work by the Tavistock Institute. f In 
the American context, the need of a federal agency to adapt some 
of its policies and goals to local values if it is to get the local 
support it needs is brought out in Selznick's TVA and the Grass 

* Reinhard Bendix, "Bureaucracy, the Problem and its Setting," The 
American Sociological Review, 12:493-507 (1947); and Stephen A. 
Richardson, "Organizational Contrasts on British and American Ships." 
Administrative Science Quarterly 1:189-207 (September 1956). 
f A. K. Rice, Productivity and Social Organization: The Ahmedahad 
Experiments (London: Davis Pubhcations, Ltd., 1958). 

parsons' theory of organizations • 237 

Roots."^ The reader will note that the first two studies cover the 
eflFect of culture on coordinating decisions (Gi); while the other 
two cover the effect of culture on the acquisition of labor (Al) and 
the determination of goals (Gg). Existing studies of organizations 
can be fitted into Parsons' categories with some ease. 

Values are not the only way in which environment and organiza- 
tion are related. Equally suggestive is Parsons' emphasis on the 
fact that all organizations have a "product," that all need support 
from the environment, and that economic organizations are an 
extreme and limiting case in that they obtain such support almost 
entirely through the sale of their product. 

In his latest papers Parsons f has made this comparative tend- 
ency even more pronounced and systematic. He lists five forces 
which, compounded, result in systematic differences between 
organizations: (1) differences in level (whether one is studying 
the high school— technical level— or the Office of Education— insti- 
tutional level); (2) differences due to technology (differences 
between school and factory due to technical differences in the 
processes involved); (3) differences due to varying exigencies of 
procuring and disposing of whatever the organization needs (a 
church obtains different things, and obtains them in different ways, 
from an economic organization); (4) differences due to location in 
different functional sectors of society; (5) differences in "articula- 
tion between levels"— thus, professional personnel are supervised 
differently in the military from the supervision they receive in 

As an example of the effect on the recruitment of personnel of 
the requirement to keep community support. Parsons suggests that 
because mental hospitals depend for their success on an extreme 
degree of public confidence, there will be a tendencv to have 
psychiatrists as top administrators— the post most "visible" to the 
public— since the public has most respect for doctors. Unlike Whyte, 
I do not regard it as a serious deficiency that, in fact, fewer 

* Philip Selznick, TV A and the Grass Roots: A Study in the Sociology 
of Formal Organizations (Berkeley: University of California Press, 

f Parsons, "Some Ingredients of a General Theory of Formal Organi- 
zation," in Halpin, op. cit. 

238 • Henry A. Landsberger 

psychiatrists may be heads of mental hospitals than professors at 
the head of universities. Parsons is necessarily confined, at any one 
point, to talking about a single cause in a situation in which there 
are other forces as well. Other tendencies in the situation, of an 
opposite kind,— in the case of the mental hospital, the relative 
greater scarcity of psychiatrists— may well overcome in the final 
outcome the single tendency about which Parsons talks. This does 
not mean that the tendency is absent. 

The aspect of organization not adequately covered even by 
Parsons' comprehensive scheme is that of technology. Despite 
repeated references to the fact that among the adaptive problems 
of organizations are requirements to adapt to "technological exigen- 
cies" and despite references to the effect on the organization of 
the nature of its goal. Parsons' theoretical system cannot system- 
atically handle the influence of technology on organizational rela- 
tionships. He gives examples of the influence of technological 
"exigencies" on authority structure, etc., but these are ad hoc 
insights and not deduced from his system. Altogether, he has little 
interest in technical activities as such: ideally, they are routinized, 
and become of interest to the administrator only "when something 
'goes wrong.' " The effect of technology on human relations has 
recently stimulated a good deal of research, though it is in my 
opinion too early to state how important a variable technology is. 
Nor would its importance ever preclude the simultaneous opera- 
tion of the other kinds of variables in which Parsons is interested. 
However, Parsons' theory is here faced with the same diflSculty as 
are all sociological theories when attempting to integrate nonhuman 
variables, whether they be technology, geography, or climate. Vari- 
ables of this kind have first to be translated into social variables 
before they can be systematically handled by a social theory, and 
even then, their incidence and strength is bound to be random and 
unpredictable, however important their effect may actually be. 
However much the four system problems and their sub-problems 
may cover they do not allow very well for problems of this kind. 

This particular deficiency highlights that the specialist in any 
one of the social sciences will always have a legitimate doubt about 
the utility of a "general theory." By definition, a general theory is 
designed more to show the equivalence between one field and 

parsons' theory of organizations • 239 

another than to shed light on the problems unique to a single field. 
It is in the latter that the specialist is likely to be interested: he 
wonders whether it is worth his while to understand a theory which, 
as Parsons clearly recognizes, "has involved bringing together con- 
siderations from a variety of specialties in ways which the respective 
specialists would seldom think appropriate or useful." * 

The Occupational Contract and System Interchanges 

The occupational contract is the most important mechanism 
through which social systems, and organizations in particular, solve 
the two adaptive problems of obtaining labor and entrepreneur- 
ship. In Parsons' hands, this contract allows for a very complicated 
set of exchanges; first, because a triangular relationship is involved 
(the organization strikes a contract with the individual's person- 
ality system as well as with the household system to which he 
belongs) and second, because each of these two contracts has the 
usual four sectors, of which the wage-service exchange is merely 
one, namely the G component (Figure 11). 

The other three exchanges are: (1) the "A" bargain, in which 
the worker, both in his family role and as a personality, gives the 
organization more or less right to intervene. He may adapt both 
himself and his family to the exigencies of work in return for 
"credit" in the form of greater or lesser assurance that he will con- 
tinue to hold his job. The likelihood of an organization exercising 
its right to hire and fire is therefore involved, and according to 
Parsons, the worker's job security literally aflFects his credit standing 
in the community. While Parsons himself is probably not aware 
of it, there is a paradoxical example of this relationship during 
steel strikes, when local stores extend almost limitless credit to 
strikers, knowing that they will ultimately return to the jobs await- 
ing them. There is a parallel between this A exchange and the A 
exchange at the larger, societal level, in which the economy gets 
credit (capital) in exchange for the right to intervene by those 

* Talcott Parsons, "General Theory in Sociology," Ch. I, pp. 3-38 in 
Sociology Today: Problems and Prospects, Robert K. Merton, Leonard 
Broom, and Leonard S. Cottrell, Jr., (eds.) Basic Books Inc., New 
York, 1959, p. 36. 

240 • Henry A. Landsberger 




G Labor Services 



( Purchasing power) 

Rights to intervene: 
Attitude to authority in 
organization: acceptance 
of authority and executive 

Credit creation: 
Credit standing: capital 
"making advances" to labor 

I Influence: 

Reputation as good 

Contingent support: 
Responsibility for household 
welfare: fringe benefits 

Defined in terms of 
security of household- 
entrusted to the economy 

Moral approval: 
Defined in terms of values 
of economic rationality: 
production for standard of 

Figure 11. Labor market. 

extending it. Both the household as a system and the personaUty 
as a system will attempt to limit the extent to which they adapt 
themselves to the occupational role and will seek some adaptation 
on the part of the organization. 

The kinds of intervention and adaptation which may be involved 
are numerous. Included are such items as the extent to which the 
breadwinner is expected to worry about his job while at home (an 
intervention in the household), and the extent to which the organi- 
zation allows him to bring his family worries to work. 

A second exchange embodied in the contract is the integrative 
(I). Mutual support and loyalty and an exchange of prestige are 
involved. As a personality, the worker gets prestige and therefore 

parsons' theory of organizations • 241 

psychological support from his occupational role, while reflected 
prestige, fringe benefits and security of employment are forms of 
diffuse support for the household. The worker in turn may give 
loyalty and support to the organization. He adopts a more diffuse 
responsibility for the job than specified by the wage-service ex- 
change, as well as giving the employing organization a good name 
in the community. It will be remembered that the I sector is char- 
acterized by the pattern variable of diffuseness among others. 

Finally, the latency aspect of the contract concerns a mutual 
recognition of each others' values, tangibly expressed in the other 
three areas. The worker and the household accept effective pro- 
duction and being "a good worker" as a value while the organization 
recognizes a moral obligation to pay good wages, to recognize 
family responsibilities. 

Because American values emphasize the A sector in general, 
occupations— which are by definition adaptive— tend to get more 
than they give. Parsons is aware of the tension to which this gives 
rise for all personalities except those with high A (achievement) 
motivation. The household, too, puts more into the contract than 
it receives. Families will be disrupted because occupational success 
and achievement are so important. 

Trade unions are mechanisms which Parsons sees as preventing 
or correcting at least some of the tensions created by the economy 
and the occupational system. He sees the union as primarily con- 
cerned not with the wage-service component of the contract, but 
with the A component: the worker's acceptance of authority and 
discipline as a person, and his willingness to adapt family life to 
the job. This is the sector which produces most tension (ES 148). 
It is the regulation of the organization's right to intervene in the 
individual's life that is the union's prime function. In addition, it 
has a considerable effect on the worker's attitude to work and 
himself (I and L). Through protecting him, giving him greater 
self-confidence, and the feeling of being respected and approved 
(L), the union helps workers accept general conditions of employ- 

Emphasizing the union's function in restricting the employer's 
right to intervene is perceptive. Restriction of management's right 
to fire and to discipline the individual at work, with the chaos this 

242 • Henry A. Landsberger 

brings into the worker's life, has indeed been a major goal for 
unions. Parsons spoils our appreciation of his acuteness, however, 
by stating elsewhere that "it is notable that the growth of trade 
unionism in the United States has been accompanied by relatively 
little demand for managerial prerogatives." (ASQ 235) Since dis- 
cipline clearly falls into the disputed area of managerial preroga- 
tives, this is a contradiction. It is likely, however, that Parsons was 
comparing the United States with a phenomenon such as co-deter- 
mination in Germany where unions actually participate in the 
management process, and in this he is correct. In any case, it is 
interesting that he is sufficiently in touch with the world to realize 
that he cannot afford to fail to comment on as important an institu- 
tion as labor unions, and that his basic evaluation of them is a 
positive one. His analysis, while not novel, would probably not be 
rejected as basically wrong by experts in labor relations. He allows 
for the strictly wage determination function of unions in straight 
bargaining, conflict-of-interest, terms. But primarily he sees them as 
a protection of workers in a situation where the worker, because 
of his isolation and the moral approval given to industrial (adap- 
tive) activities, is in a weak position. Without such protection, the 
system would break down. This assessment of unions as a necessary 
balancing mechanism contrasts favorably with the early writings 
of the Mayo group— though Parsons has the advantage of some 
twenty years of further history on which to draw. 

The fourfold division of the contract, like the fourfold division 
of organization problems and structures, is arbitrary in detail. Could 
Parsons answer the many questions which spring to mind? Why, 
for example, is security of employment mentioned twice— once in 
the A sector, as giving credit specifically, and again in the integra- 
tive sector, as giving diffuse support to the family? Is it because 
any concrete item, such as security of employment, may analytically 
have more than one kind of functional significance? This would be 
acceptable logically; would be quite in line with Parsons' general 
modes of reasoning, and would resolve one or two other puzzles to 
which I refer below. 

Yet despite these and many other conceptual and operational 
ambiguities, there is something of considerable value in this 
analysis of the employment contract. 

parsons' theory of organizations • 243 

In the first place, the idea is an interesting one that there are 
really two employment contracts: one between organization and 
household, another between organization and personality. Though 
it seems to complicate matters, it corresponds well to the psychology 
of employment and its dilemmas: the choice, for example, between 
a personally more satisfying job and one which allows a better 
meeting of family responsibilities.* Parsons regards this link between 
the economy and personality via the occupational contract as a 
major theoretical contribution because "The paradigm for the con- 
tract of employment is thus the main framework for the transition 
between the sociological and the psychological analysis of indi- 
vidual motivation in the occupational role." 

Parsons regards the entire idea of a contract, with multiple di- 
mensions along which exchanges can occur, as a fruitful way of 
thinking about the relationship between, and the integration of, any 
two systems or any two subsystems. For the occupational contract 
not only integrates an individual household and an individual 
organization, but through this process enables one to examine the 
state of integration of the larger systems of which both are a part: 
the economy as a whole, with the whole of the latency sector 
of society. Thus, two systems of society, as well as various levels 
within those systems, interpenetrate each other via "the same con- 
crete behavior process." (ES 115, note I) 

As a theory of wages. Parsons' ideas need to be compared with 
the present state of wage theory in sociology and economics. Both 
disciplines are aware that they need each other. Neither, however, 
has found its way to the other. Whyte's Money and Motivation f 
may be taken as an example, since it is the most outstanding socio- 
logical ejBFort in the field of wages, yet ends with an unfulfilled 
promise. His is a vivid description of individual and group reac- 
tions to wages, particularly to incentive systems. Various readings 
in this book stress that wages do not exhaust what people ex^pect 

* See, for example, Charles Walker and Robert H. Guest, The Man 
on the Assembly Line (Cambridge, Mass.: Harvard University Press, 

f William F. Whyte, Money and Motivation: An Analysis of Incen- 
tives in Industry (New York: Harper & Brothers, 1955). 

244 • Henry A. Landsberger 

to get from a job: they like to control the pace of their work (in 
Parsons' terms: limit the right of others to intervene), and they ex- 
pect to maintain prestige. Wages, being a "concrete item," may stand 
as much for prestige as for purchasing power. But the book stops 
short of spelling out systematically what the worker seems to ex- 
pect from his contract apart from wages as purchasing power. 
Prestige is mentioned, but is not conceptualized as a possible ele- 
ment in an exchange. More importantly, neither Whyte nor others 
have drawn attention to the fact that the organization, too, may ex- 
pect more than specific performance of tasks. 

Parsons realizes, of course, that the importance of these various 
elements varies from occupation to occupation— indeed, his theory 
is intended to allow for this kind of variation quite systematically. 
Thus, the executive's family is expected to adapt a great deal to 
his job— in return for which he has greater job security than a blue 
collar worker. The executive is expected to take on a great deal of 
diflFuse responsibility, and his high salary, according to Parsons, 
symbolizes this rather than the value of his specific services. The 
doctor gives the patient (his employer) no right to intervene but 
expects the patient to adapt to him, and also to give him a great 
deal of diflFuse loyalty (no shopping around). In return, the doctor 
renders his services for less payment than they are worth to the 
employer (sliding scale for the poor). Organizations in the L sector 
of society (e.g., educational institutions) are not on a money -for- 
product basis as are organizations in the adaptive sector of society. 
Hence, they cannot pay for their occupational roles in the form of 
high wages. Instead, they give prestige and tenure. 

Once again, the empirical assertions may not be fully accurate, 
and the fit with the four system problems is arbitrary and ill de- 
fined in many instances. It is a matter of some doubt whether the 
intellectual's prestige is high: although one would not expect it to 
be very high, even using Parsons' method of analysis, since Ameri- 
can society stresses A, not L, values. But the assertion about the 
greater job security of executives is definitely open to query, and 
Parsons himself seems to have recognized this. In one of his latest 
statements he contradicts himself by saying that the executive's 

parsons' theory of organizations • 245 

high remuneration and low job security are symbolically significant 
of the entrepreneur's position as risk taker.* 

Despite ambiguities, these categories are important ones. They 
point to systematic comparisons between situations which had pre- 
viously been thought noncomparable, and the problem of making 
them operational and subject to empirical investigation should not 
be insurmountable. Such matters as the household's receipt of 
prestige, security, and approval are covered by items in existing 
morale surveys, and there are many other methods for getting at 
them. Little has been done but much could be done by way of 
asking employers whether they get what they think they are paying 
for. The evaluation of the loyalty and responsibility of their work 
force would make a great deal of sense to employers, who 
spontaneously talk in these terms. I well remember a manager who 
had moved from London to the North of England and bemoaned 
the fact that, while Londoners were insubordinate (low on A: "right 
to intervene") they see a job through even if instructions do not 
cover certain contingencies (taking "diflFuse responsibility"). "Up 
North," the worker— while more docile— would either give up or 
spoil the job by adhering to inapplicable instructions. 

Others of Parsons' predictions are likewise capable of measure- 
ment, and are suggestive. For example, one way of dividing tech- 
nical occupational roles (roles other than entrepreneurial ones since 
these have general, diffuse, responsibility for the organization by 
definition) is according to whether "the technical function involved 
is in line with the function of the organization in the society, e.g., 
the physician in a hospital . . . , or whether the technical function is 
auxiliary to the organization's primary function, e.g., the physician 
in the school system." (ES 118) Such a distinction suggests that it 
may be erroneous to think of an occupation as having a fixed level 
of power and prestige. The power and prestige of an occupation 
may be affected to some extent by whether or not it is primary to 
the sector in which it is found. 

Economists, like sociologists, have not succeeded in formulating 

* Parsons, "Some Ingredients of a General Theory of Formal Organi- 
zation," in Halpin, op. cit. 

246 • Henry A. Landsberger 

a convincing theory of wages. Their fate is a highly instructive one 
for all of us and for Parsons' scheme in particular. The history of 
wage economics is, very crudely, as follows. The concepts of mar- 
ginal productivity and the theories of supply and demand held that 
workers would be paid according to what they were worth. Em- 
ployers would pay less and less if they were asked to hire more 
and more labor, since, according to the theory, productivity de- 
clines as more workers are hired. Equilibrium was reached where a 
downward sloping demand-for-labor curve intersects an upward 
sloping supply-of-labor curve. This was an equilibrium theory in 
the full sense of the term, since full employment was assured by 
the fact that workers would reduce their price in order to find em- 
ployment. Indeed, the level of wages was that at which everyone 
who wanted to do so could find employment. Inter-industry and 
other wage differentials were likewise settled by supply and de- 
mand. People moved to areas, to industries, to occupations, or to 
levels of skill which offered unusually high wages. 

This model does not correspond to reality in many important 
respects. Wages and the level of employment are not determinants 
of each other, for wages do not move down at times of unemploy- 
ment to absorb the unemployed. So how can one account for the 
existing level of wages? As between industries, jobs, and areas, etc., 
strange wage differentials likewise persist without influx of labor. 
Disillusioned with theory, economists have recently been engaged 
in discovering and describing what, empirically, was going on, 
leaving for later the attempt to explain what might be found. Re- 
gardless of explanation, what has happened to wage differentials? 
When do people move and to what extent do they move? Essen- 
tially, labor economists are asking the very modest, but very im- 
portant, questions: What are the facts and what, at a very humble 
empirical level, are the empirical relations between the facts? 
Speculation about underlying variables and forces and the build- 
ing of a new model can come later. (This, of course, is an exag- 
gerated description: labor economists are continually formulating 
"middle range" theories, but are having to discard them for lack 
of general validity. ) 

The lesson is that the difficulty is not one of thinking up variables, 
nor of finding empirical indices of them. Indices of wage rates, of 

parsons' theory of organizations • 247 

mobility, and of employment, while imperfect, are as good as they 
are in any field in the social sciences. Nor is the difficulty that of 
gathering data. The real difficulty is that the empirical data "do 
not make sense." No one has hit on the actual, mathematical rela- 
tionships between the variables which would actually account for 
the data, hence no one can set up a model. 

This has considerable relevance for Parsons. For in principle an 
economist might grant that Parsons has now supplemented the wage- 
service bargain by a systematic specification of at least some of 
the other elements which are bargained for, and that this has been 
done in terms which have general applicability. He might also grant 
that operational indices of these elements can be obtained. The real 
difficulty, however, will arise over explaining why just so much 
contingent loyalty is exchanged for just so much right to interfere 
in one occupation, and more in another. Parsons will probably try 
to work with four variables simultaneously (or with eight if the 
personality contract is included) balancing them ofiF against each 
other. The economist, on the basis of his experience, might predict 
that Parsons will have trouble finding the equations, that his diflS- 
culty will be sixteen or sixty-four times as great as that of the 
economists. Parsons would, of course, reply that whereas one can- 
not find equations for just the wage-service bargain, one can do so 
if the balancing eflFect of the other three elements are taken into 
account. He would have to concede, however, that as yet he has 
not proposed a set of mathematically specified relationships— not 
even a set of real hypotheses. 

It is apparent that there are grave deficiencies in Parsons' theory. 
There is the lack of clarity in underlying purpose— whether Par- 
sons' scheme is explanatory and descriptive of what is, or sets pre- 
conditions and minimal conditions for stability, or whether it is an 
ideal type. There is the conceptual vagueness of terms such as 
"sector" and the absence of tight links to empirical indices of con- 
cepts such as "adaptation." There is the failure to specify quantities 
or even, except in rare instances, to hypothesize functional rela- 
tionships, leaving the theory at the level of a set of categories and 
mechanisms rather than predictions. Indeed, it remains to be seen 

248 • Henry A. Landsberger 

whether these concepts lend themselves to being converted into 
variables about which hypotheses can be formed. There is, finally, 
the failure to cover certain causal variables systematically. When 
judged against the goal which Parsons has set for himself, one can- 
not but feel that the deficiencies at present exceed the achieve- 
ments, and continued progress is by no means assured. 

An alternative way to evaluate Parsons' work is to judge it against 
previous progress in the area in which he has chosen to work— the 
construction of a general theory. For most fields in the behavioral 
sciences— especially psychology— such an attempt is not of outstand- 
ing importance. For sociology and organization theory this aim is 
of great importance. Without implying a Comtian hierarchy of 
sciences or a psychological reductionism, it is apparent that sociology 
and organization theory need other disciplines more obviously than 
psychology does. Parsons' scheme alone starts out from the premise 
that levels of analysis (sociology versus psychology), and func- 
tional areas (sociology versus economics and politics) have to be 
related to each other and that the issues with which each deals 
have substantive similarities. Regardless of the merits of the scheme, 
this intention to be comprehensive and to jump disciplinary borders, 
and the seriousness with which Parsons pursues his intention, are 
greatly to his credit. 

As for the merits of the scheme, his concepts refer to important 
phenomena, certainly in the field of organization theory. No theory 
of organization could fail to cover, yet none exists which does 
cover, the acquisition of resources, the strain to reach a goal, in- 
ternal integration, larger social values, and— most important— the 
circular influence of each on the other. Parsons' analyses of, and in- 
sights into, specific problems fit quite well into these categories 
while at the same time being congruent with the formulations of 
some of the most advanced writers in the field. This is a remarkable 
achievement when one remembers that these categories were set 
up to analyze completely different systems also, and that they do 
so with at least a modicum of success. Parsons has advanced a 
very modest degree toward covering common system problems, in 
a field in which no other modest advances exist. Finally, Parsons' 
concepts and their relationship pose new problems for research, 
often in key areas that have been overlooked. I have found no 

parsons' theory of organizations • 249 

more difficulty in thinking of these concepts in operational terms 
than of any other set. In view of the wide range of knowledge which 
Parsons demonstrates, and the elegance and subtlety of his argu- 
ment, I find it difficult to be anything but respectful of his con- 
tribution while fully recognizing its present limitation. 



William Foote Whyte 






A critique of the theories of Talcott Parsons as they 
apply to organizations must begin with the recognition 
that Parsons has devoted very little attention to organi- 
zations and organization theory. I am sure he would be 
the last man to say that at this point he has presented an 
adequate theory of organization. Nevertheless, he is under- 
taking to develop a general theory of society, and such 
a general theory should provide us with some useful 
guides for the examination of organizational life. We may 
therefore legitimately ask: Insofar as organization theory 
goes, is Parsons on the right track? My most general an- 
swer can be given in one word: No. 

I found these three main difficulties with the Parsons 
effort in the field of organization theory: 

1. He is concerned primarily with boundary relations, 


the relations between the organization and society. He gives Httle 
attention to behavior v^ithin the organization. 

2. His concepts do not hnk up with observable data. 

3. He omits a number of elements that seem to me essential for 
building organization theory. 

I shall take up these objections in turn. 


In the first of two articles on "Suggestions for a Socio- 
logical Approach to the Theory of Organization," Parsons states his 
point of view in these words: 

The main point of reference for analyzing the structure of any 
social system is its value pattern. This defines the basic orienta- 
tion of the system (in the present case, the organization) to the 
situation in which it operates; hence it guides the activities of 
participant individuals. (ASQ 67) 

Parsons calls this focus on the orientation of the system to its 
situation the "cultural-institutional" point of view. He recognizes 
that this is not the only way of viewing organizations. A second 
possible approach is one which involves a "group" or "role" point 
of view, 

. . . which takes sub-organizations and the roles of individuals 
participating in the functioning of the organization as its point 
of departure. (ASQ 67) 

In these articles Parsons chooses to limit himself primarily to a 
discussion of the "cultural-institutional" point of view. Such a limita- 
tion is legitimate enough for one theoretical essay. We can hardly 
expect a theorist to cover everything at once. However, students 
who peruse other Parsons' articles on organizations will find him 
still pursuing the cultural-institutional point of view. The group 
or role point of view continues to receive scant attention. Thus, the 
student of organization who wishes to profit from Parsons' writings 
should expect to find help primarily in analyzing the stance an 

252 • William Foote Whyte 

organization takes to the surrounding society and not in terms of 
guides to the internal dynamics of the organization. Let us see 
what hght he does throw on these boundary relations. 

This approach enables Parsons to distinguish between business, 
military, and university organizations. Some might say that the 
distinctions are so obvious that the contribution is hardly worth- 
while. This is not my criticism. Often, in social science, it is neces- 
sary to state systematically the obvious. I am rather concerned with 
the breadth of the generalizations that Parsons makes. 

For example, let us not just compare business organizations and 
military organizations in general. Let us take as examples two units 
in each field: (1) a mechanical maintenance organization at a major 
Air Force base, (2) a large clerical office in Air Force headquarters, 
(3) a mechanical maintenance unit of a major commercial airline, 
and (4) a clerical department in the home office of the airline. If 
we were making comparisons among these four, which pair would 
seem more similar and which pair would seem more diflFerent? I 
suspect that if we went in and observed these four units, we would 
find more organizational similarities concentrated in units doing the 
same type of work than within the two units having the military or 
the business classification. 

This is not to say that there are no significant general differences 
among businesses, military organizations, and universities. Of course 
there are, and Parsons has pointed to some of these. My illus- 
tration is simply designed to show that only a few crude general 
statements can be made at this level. If we push beyond this very 
high level of generality, we soon reach a point where our state- 
ments are more misleading than factual. 

I have pointed out that there are common elements among these 
large organizational categories. It is also important to note that 
there are great differences within any given category. My own 
research experience has been confined primarly to studies within 
private industry, but this has covered industries as diverse as 
restaurants, hotels, petroleum, steel fabricating, plastics, glass, and 
aluminum. No theorist can afford to lump such industries together. 
This is not to say that there should be one organization theory for 
the petroleum industry, another for the restaurant industry, and so 
on. It is to say that the theory that ultimately emerges will have 


to take this diversity into account. Parsons treats private industry 
as if it were all of one piece. 

Let us see how Parsons fares as he examines one type of organi- 
zation—in this case, the mental hospital,* At the outset, Parsons 
seeks to explain (a) why doctors dominate lay administrators in 
mental hospitals, and (b) why doctors are more often appointed to 
head those institutions than are professors appointed to the top 
university positions. 

Parsons begins by pointing to the values placed by society upon 
the activities of the mental hospital. Health is certainly a key value, 
and the doctor is identified with the curative process, whereas the 
administrator is not. This strengthens the hand of the doctor as 
against the administrator. 

So far so good. But the same logic would seem to apply to the 
general hospital, which is also identified with the health value. 
Whereas nearly all mental hospitals have psychiatrists as chief ad- 
ministrators, the general hospital is as likely as not to have a lay- 
man as administrator. Why this difference? 

The author gets into further difficulties with his mental hospital- 
university comparison. He states it in this way: 

Correspondingly, the relative frequency with which non-academic 
people are made university presidents is probably associated 
with the greater diversity of the total faculties of a university 
(including professional faculties) as compared with the profes- 
fessional staff of a hospital. f 

What frequency? Logan Wilson makes this comment: 

There is a widespread but mistaken notion that stands in need 
of correction with reference to the backgrounds of college and 
university presidents. One well-known commentator has stated 
that not more than a third of the group is derived from the 
ranks of professors. An investigation of the occupational experi- 
ence of presidents of the thirty leading graduate institutions in 

* Talcott Parsons, "The Mental Hospital as a Type of Organization," 
in M. Greenblatt, Daniel Levinson, and Richard H. Williams, eds., 
The Patient and the Mental Hospital (Glencoe, 111.: The Free Press, 
1957), pp. 108-129. 
f Ibid., p. 119. 

254 • William Foote Whyte 

this countiy reveals that only two of the major executives have 
had no professorial experience.* 

Is Parsons really trying to explain the difference between 28 out 
of 30 and 30 out of 30? That seems like a fine point indeed. Of 
course, it might be argued that Wilson's figures are now about 
twenty years old and limited to tliirty cases, but Parsons presents 
no figures at all, early or late. His use of the phrase, "relative fre- 
quency," suggests that he himself is a victim of the "widespread 
but mistaken notion" that Wilson sought to correct years ago. 

Parsons appears to get on firmer ground as he views in broad 
outline the nature of mental hospital activities. He asks why it is 
that mental hospitals have a tendency to concentrate upon cus- 
todial care, often at the expense of therapy. Having already noted 
the tendency of doctors to gravitate toward the top prestige posi- 
tions in administration, he draws attention to the general condition 
of shortage of funds facing these institutions. This means that the 
institutions tend to spend a disproportionate share of these scarce 
funds on high-priced medical personnel for administrative posi- 
tions and do not have the resources to employ enough medical 
personnel to carry out effective therapeutic programs. This condi- 
tion tends to leave the patients largely to the care of nurses, aides, 
and attendants. They naturally are inclined to look upon patients 
in terms of how much trouble patients cause them. A passive or 
cooperative patient is to be preferred to one who is obstreperous, 
even if the passive patient is making no therapeutic progress. 

This analysis seems useful and sound as far as it goes. In fact, 
the points stated here have been recognized at least implicitly by 
some innovators in the field of mental health. The concept of 
milieu therapy is one answer to this problem. Psychiatrists have 
recognized that patients' progress depends upon their total experi- 
ence in the institution. Since they will inevitably spend most of 
their time with fellow patients, nurses, aides, and attendants, some 
psychiatrist-administrators have sought to train the nonmedical per- 
sonnel in a therapeutic approach and have sought to develop group 
activities among patients. The omission of these developments is 

* Logan Wilson, The Academic Man (New York: Oxford University 
Press, 1942), p. 85. 


not a criticism of the theory in itself, for the conception of the 
miHeu therapy fits in quite well with the analysis of the deficiencies 
of the traditional mental hospital provided us by Parsons. 

Mixed in with some questionable generalizations and one error 
of fact, we find in this article some useful ideas. It is hard to call 
them novel ideas because they seem so familiar to anyone who has 
read casually in the mental health literature. But let us assume that 
the ideas are both new and important. Even so, we must note the 
limited range of their applicability. 

Parsons is seeking to explain how the societal environment afiFects 
the selection of key personnel and the development of activity pro- 
grams in the mental hospital. But suppose we wished to compare 
the impacts of two hospital administrators, each on his own hospi- 
tal? Or the eflFects of two different patterns of organization in two 
mental hospitals? Since the stance of the mental hospital in rela- 
tion to its broad societal environment is the same in each of these 
cases, the Parsonian approach can only deal with similarities within 
the same type of institution or differences between different types 
of institution. We look in vain to Parsons for guidance in examining 
differences in behavior, within the same type of organization. 


Elsewhere I have stated the following criticism: 

Parsons expresses a great interest in "action," and indeed the 
word is in the title in three of his books. This would lead one 
to think that Parsons was interested in the actions of people— in 
what they actually do. In this case, the appearance is deceptive. 
Parsons is instead concerned with "the orientation of the actor 
to the situation." In the world of Talcott Parsons, actors are 
constantly orienting tJiemselves to situations and very rarely, if 
ever, acting. The show is constantly in rehearsal, but the curtain 
never goes up. Parsons focuses on the process whereby the in- 
dividual sizes up his social environment and makes up his mind 
about what he might do. At this point he stops. It is precisely 
at this point that some of us wish to focus our attention.* 

* William F. Whyte, Man and Organization (Homewood, 111.: Richard 
Irwin & Company, 1959), pp. 40-41. 

256 • William Foote Whyte 

Apparently there is no disagreement between Parsons and myself 
regarding what he is attempting. In his Cornell discussion of the 
question put to him by critics, Parsons had this to say: 

What I have been calling the "Theory of Action" is clearly not 
a theory of behavior in the more immediate sense, particularly 
concerned with the physical movements of organisms in relation 
to the physical environment, including the processes of physical 
production in the partly economic sense. It is rather a theory 
that is concerned with the analysis of certain mechanisms which 
control behavior in this latter sense, which therefore in the old- 
fashioned, behavioristic sense, are not visible, not immediately 
observable, which I think in the organism operate overwhelm- 
ingly in the brain. 

Parsons recognizes that there is a gap between his system of 
concepts and observable data. He puts it this way: 

We have pointed out that the behavioral units, which have to 
be the units of empirical observation, in all probability cannot 
be the system units. This is essentially to say that it is unlikely 
that the theory of action will be able to do without the use of 
intervening variables. (WP 108-9) 

What then are to be the links between the "system units" and ''be- 
havioral units"? In a collaborative effort with R. Freed Bales, Par- 
sons claims to have found the necessary links. For his interaction 
process analysis method, Bales has divided all observable behavior 
in discussion groups into twelve categories. An exposition of the 
Bales schema would take up too much space for the present dis- 
cussion, and I will assume that readers are generally familiar with 
it, as it has indeed appeared widely in the literature. Parsons now 
proceeds to argue that some of his own concepts link up with those 
of Bales. Since the Bales categories organize the direct observa- 
tion of behavior, this would seem to bring Parsons down to earth. 

Does this combination enable us to advance our analysis of group 
behavior? Or is it just a feat of translation— from the clear to the 

Regarding the link between the concepts of Bales and Parsons, 
we have only the affirmation that the two sets of concepts fit to- 
gether. However plausible the argument may seem, we are given 


no way of testing the fit nor any way of relating propositions derived 
from one set of concepts with those derived from the other. Even 
if we assume that the fit of concepts is as good as Parsons claims 
it is, he has not gone on to demonstrate how these linkages enable 
him to go beyond the very useful analysis of group behavior that 
Bales carries out with his own conceptual tools. 

The absence of any propositions at all is noteworthy at this point. 
It is only when Parsons goes on to talk about the equilibrium of 
social systems that he seems to be concerned with propositions. Fol- 
lowing the model of classical mechanics, Parsons derives the fol- 
lowing laws on the equilibrium of social systems: 

1. The Principle of Inertia: A given process of action will con- 
tinue unchanged in rate and direction unless impeded or 
deflected by opposing motivational forces. 

2. The Frinciple of Action and Reaction: If, in a system of 
action, there is a change in the direction of a process, it will 
tend to be balanced by a complementary change which is 
equal in motivational force and opposite in direction. 

3. The Principle of Effort: Any change in the rate of an action 
process is directly proportional to the magnitude of the moti- 
vational force applied or withdrawn. 

4. The Principle of System-Integration: Any pattern element 
(mode of organization of components) within a system of 
action will tend to be confiimed in its place within the sys- 
tem or to be eliminated from the system (extinguished) as a 
function of its contiibution to the integrative balance of the 
system. (WP 102-3) 

These propositions give the impression of concreteness, without 
the substance. To make them applicable, we really need to know 
(a) what are the units of observable action to which the proposi- 
tions apply? and (b) how are these units quantified? Since, as 
usual, Parsons gives us no hint regarding the connection between 
his principles and data, we are simply left up in the air as to the 
utility of the propositions. 

I cannot leave this point without calling attention to the unusual 
nature of this performance. Here we have a major social theorist 
stating propositions whose meaning is completely unclear and then 
going on to other matters without making the slightest effort to re- 
late these propositions to the world of real behavior. 

258 • William Foote Whyte 

I have looked elsewhere in Parsons for links between his con- 
cepts and the world of behavior,, with little more success. At various 
points in his writings he deals with terms such as legitimation, in- 
stitutionalization, allocation, decision-making, integration, authority, 
staflF and line, and so on. Since at least some of these terms are 
commonly used by other theorists of administrative behavior, we 
might assume that we are coming closer to the level of observable 
behavior than is usually the case with Parsons. This again seems to 
be an illusion. 

For example, in his article on the mental hospital, Parsons makes 
these statements: 

On the one side, the powers and authority of administration 
must be legitimized. . . . 

The prime base in the structure of organization for the accept- 
ance of the consequences of policy decisions, on the other hand, 
is the institutionalization of authority.* 

When something is being "legitimized" we assume that the verb 
refers to a process that is going on. Parsons does not tell us what 
this process is. We are simply left to assume that, if people accept 
authority as legitimate or proper, somehow it has been legitimized. 
Parsons is therefore referring to a state of sentiments or attitudes 
among members of an organization rather than to any process. 

The same criticism may be offered regarding the "institutionali- 
ation of authority." Authority has become institutionalized when 
people accept authority, but we get no indication of how institu- 
tionalization takes place or of how to know when it is or is not tak- 
ing place. 

Even the word "authority" gives a misplaced sense of concreteness. 
We all think we know what authority is, but when we get to the 
task of actually studying an organization, we have great difficulty 
in giving the term any operationally acceptable meaning. Are we 
talking about the oflBcial and formal theory of the organization re- 
garding who has authority and who has not? Or are we dealing 
with the patterns for initiating action that are to be observed within 
the organization? As we all know, official determinations regarding 

Parsons, "The Mental Hospital," pp. 124, ]25. 



the allocation of authority are often quite at variance with the be- 
havior to be observed in the organization. So we have to ask again, 
what does Parsons mean by "authority"? Like other questions de- 
signed to link concepts to data, this one receives no answer. 

The same problem arises when Parsons refers briefly to line and 
staff in his first Administrative Science Quarterly article. He de- 
fines staff as 

. . . usually some kinds of experts who stand in an advisory 
capacity to the decision makers at the various levels, but who 
do not themselves exercise "line" authority. (ASQ 69) 

It is indeed an unusual case where the meaning of one term is 
defined by another term whose meaning is also in question. But 
there are more serious difficulties involved in this definition. 

This definition of staff people as advisory might have been rea- 
sonably adequate for organizations in which nearly all of the per- 
sonnel were concerned in the central production activity and just 
a few experts constituted the staff. Now most large-scale organi- 
zations have developed large staffs involved in carrying out their 
own specialized activities. Do they have authority in relation to the 
so-called line organization? That depends upon what we mean by 

For example, we may find the industrial relations department 
setting up and administering a wage and salary administration 
scheme. To be sure, the scheme itself may have to have the ap- 
proval of some higher level functionary, but once the scheme is 
instituted, the relevant personnel men make the decisions in this 
sphere. They have the power to approve or disapprove wage and 
salary recommendations, in terms of this scheme. 

In the handling of union grievances, we often find that an in- 
dustrial relations man is actually making the decisions. In negoti- 
ating a union contract, the industrial relations man is often the chief 
negotiator and chief decision maker. To be sure, he generally has 
to get approval from top management as to the final terms he will 
be able to offer, but often he does not just ask the top officials what 
they are prepared to give. He tells them what, in his judgment, will 
be necessary to offer in order to reach a settlement. 

In fairness to Parsons, it should be added that in a later article. 

260 • William Foote Whyte 

he does indeed begin to point out the diflBculties with current con- 
cepts of staff and line. In discussing school administration, he points 
out that the teachers do not simply carry out what the adminis- 
trators tell them to do. The teachers' 

. . . position cannot be a simple "line" position. Nor, indeed, is 
it adequate to assign them to the "staff" and say that their 
function is to "advise" the "lay" executive. This implies that it is 
the executive who really makes the decisions. But this is not 
correct. The technical expert must, in the nature of the case, 
participate in the technically crucial decisions.* 

This is a worthwhile beginning. Furthermore, the contrast be- 
tween the staff-line statements in the earlier and later articles sug- 
gests one of the difficulties which has plagued Parsons' efforts at 
organizational theorizing. The earlier work suggests a Parsonian 
assumption that he will get his feet firmly planted on the ground 
when he links his concepts up to those which have been tradi- 
tionally used in administrative theory. A further examination of the 
staff-line problem is suggesting to him that these traditional con- 
cepts are not firmly grounded. Where then should Parsons look in 
order to ground his theories? 

That question suggests a more general consideration of the re- 
lationship between Parsonian concepts and observed behavior. Let 
us look particularly at the relationship between action and the ac- 
tor's orientation to the situation. Parsons seems to assume that, if 
we know the actor's orientation to the situation, we can predict 
how he will act in that situation. (Otherwise, how justify such an 
interest in orientation?) 

Before we examine that assumption, we should note that it de- 
pends upon our ability to determine the actor's orientation to the 
situation. By what methods are we to do this? On this point, Par- 
sons gives us very little guidance. Presumably the actor orients him- 
self first in terms of the values of the society pertaining to the situa- 
tion he faces, but we have already noted that these values tend to 

* Talcott Parsons, "Some Ingredients of a General Theory of Formal 
Organization," in Andrew W. Halpin, ed.. Administrative Theory in 
Education ( Midwest Administration Center, University of Chicago, 
1958), pp. 40-72. 


be rather general and limit-setting rather than specifically determin- 
ing. Nor does Parsons tell us how to determine whether a specific 
individual has internalized these particular values. Anyone who 
has done field research knows that this is a most difficult problem. 
We try to determine a man's orientation at least in part from 
what he tells us about himself. Nevertheless, we often note that 
informants will say one thing and do something quite different. 
Nor is this generally a matter of deliberate deception on the part 
of informants. They often give us a normative account of their 
orientation. In telling us what they are planning to do, they tell us, 
in effect, what they feel they should do, and subsequent action 
often follows quite a different path. 

The difficulties of relating orientation to action are of course not 
news to such a sophisticated theorist as Parsons. I am simply point- 
ing out that his theory rests in important part upon some of the 
slipperiest data that a sociologist has to contend with. 

But let us make the large assumption that Parsons has some 
adequate way of determining the actor's orientation to the situa- 
tion. Can he predict actions from these data? Not at all. What any 
given individual will do depends in large part upon what others 
in the same situation will do. Thus, if we are going to predict for 
individual A, we would have to know not only A's orientation to 
the situation but also the orientations of B, C, D, and of all the sig- 
nificant others in that situation. And even that would not be enough 
because, as the situation unfolds, the process of interaction often 
leads individuals to act in ways that could not have been predicted 
from knowledge of their orientations. 

If we make the actor's orientation to the situation our primary 
focus of interest, we are operating within a field where the boun- 
daries of possible actions may perhaps be known but where the 
forces within the field are almost completely indeterminate. 

There are, fortunately, other ways of going about the study of 
organizations. Instead of concentrating our attention on orientations 
to situations, we can study what the actor actually does. This takes 
us into a field where the data can be objectively observed and, in 
some cases, measured. When we do this, we find that life is not 
nearly so indeterminate as it seems when we are wrestling with 
orientations to situations. We soon learn to recognize patterns in 

262 • William Foote Whyte 

the actions men take. It is from the recognition of such patterns 
that we can expect progress in organization theory. 

This is not to say that we should disregard any available informa- 
tion on the actor's orientation to the situation. Such information is 
highly useful when it is linked together with what we can directly 
observe. I am only saying that it is folly to base one's entire research 
strategy on the type of data most difficult to gather and subject to 
conflicting interpretations by different research men. Let us indeed 
take an interest in the subjective life of the actor, but let us anchor 
such information in the data we gather from the objectively ob- 
servable life of that actor and his fellows. 


Some of the elements omitted from the Parsonian scheme 
have already been implied in the previous discussion. This is the 
time to make the points explicit and to provide some illustrations. 

Parsons gives little if any explicit attention to the impact upon 
organizational behavior, of organization structure and the spatial 
location of people. We have been finding that the way positions 
are set up and related to each other has a most important impact 
upon behavior. As a corollary to this, we find that different struc- 
tural arrangements have different behavioral consequences. The 
same point may be made about the location of people in physical 
space. Both of these influences are illustrated in our study of the 
relations between plant manager and controller in the ABC Com- 

In this particular company, the organization structure provided 
that there be a manager and a controller in each plant, at the same 
hierarchical level. The manager reported to the production man- J 
ager in the divisional oflice, while the controller reported to the 
controller and vice president also in the divisional office. 

This particular structural arrangement seemed to be the source 
of a good deal more friction between plant manager and plant con- 
troller than we found in companies where the controller at the 
plant was subordinated to the plant manager. The conflict was re- 
ported to be particularly intense for the company's two largest 


plants which were located in the same city as the divisional head- 
quarters. In a smaller plant some three hundred miles away from 
divisional headquarters, we found a plant manager and plant con- 
troller getting along with very little friction. 

While personalities must be taken into account in any study of 
interpersonal frictions, the presence or absence of friction between 
these two positions could be accounted for to a substantial extent 
by a consideration of organization structure and spatial location of 
personnel. In the divisional office city, the plant manager interacted 
frequently with his superior, the production manager, and the plant 
controller interacted frequently with his superior, the controller 
and vice president. This meant that conflicts between the two equal 
authorities at plant level could readily be passed up the line for 
resolution at a higher level. The two men did not have to resolve 
their problems on a face-to-face basis. In the smaller plant, some 
three hundred miles distant, the plant manager and plant con- 
troller saw their respective superiors only perhaps half a dozen 
times a year. There was some communication via long distance 
phone calls, but each man had a budget for his telephone expenses, 
and he incurred criticisms from above when he exceeded that bud- 
get. Facing these spatial and budgetary limitations upon communi- 
cation with superiors, the plant manager and plant controller were, 
in effect, thrown together to work out their problems with each 
other. In the case we observed, the plant manager had become the 
dominant figure, and a smooth-working relationship had been es- 

I am not suggesting, from this case, that a given structural or 
geographical arrangement of personnel absolutely determines the 
relations among men. I am simply claiming that such objectively 
observable factors do have an important channeling influence and 
that organization theory must take them into account. 

Technology and the nature of the actual job operations provide 
another source of influence. The case cited earlier of the airline and 
Air Force clerical and maintenance departments was intended to sug- 
gest that the technology and nature of job operation could be ex- 
pected to have an important impact upon organizational behavior. 
Thus, certain similarities in technology and job operation could be 
expected to lead to some similarities in organizational behavior even 

264 • William Foote Whyte 

between parts of organizations commonly classified as lying in two 
distinct fields such as military and business organizations. As a 
corollary to this, we would expect that where we find significant 
differences in technology and job operations even within one gen- 
eral category such as business, we can expect to find significant 
differences in organizational behavior. Here again I am not claim- 
ing a determining influence but simply suggesting that technology 
and job content are sufficiently important in channeling organiza- 
tional behavior so that they must be taken into account in build- 
ing organization theory. 

In research, we are beginning to go beyond the mere statement 
that technology and job content are important. In examining factors 
leading to cohesion of work groups, Leonard Sayles * has shown 
that work groups which are relatively homogenous in job opera- 
tion, pay, level of skill, and so on, tend to stick together in their 
dealings with management more effectively than do groups which 
are heterogeneous in these respects. Homogeneity or heterogeneity 
is determined to a large extent by the nature of the technology and 
job operations within a particular department. 

The flow of work from worker to worker and from department to 
department provides another important dimension of organization 
neglected by Parsons. No one can make sense out of organizational 
behavior in a large restaurant, for example, unless he examines the 
flow of work from customer to waitress, from waitress to service 
pantry personnel, from waitress to bartender, and from waitress 
back to customer. 

The same point can be made for the factory or other types of 
organizations. For example, I once studied a foreman of a steel- 
barrel production department. At one time, he had been making 
production records and was getting along exceedingly well with 
workers, union representatives, and his superiors in management. 
Two years later, he was transferred to a lower status job because 
he was thought to be doing poorly in production, and he was ex- 
periencing mounting friction with workers and union representa- 
tives. What made the difference? As we would expect, there was no 

* Leonard Sayles, Behavior of Industrial Work Groups (New York: 
John Wiley and Sons, 1958). 


single factor in this case, but a major part of the explanation was 
found through an examination of the work flow through the de- 

Joe Walker, the foreman in this case, had never got on well with 
foremen in other departments and had always been weak in plan- 
ning his work. In the period of his success, these deficiencies were 
unimportant. At that time, the company was operating in a seller's 
market, so Joe Walker's production was confined to long runs of a 
few simple types of barrels. On each of the two lines in his depart- 
ment, he might be running the same order all week long. This 
called for a minimum of planning activity and for a most simple 
paper work operation. It also made for infrequent interactions with 
other departments. One order placed every week or so wdth the 
steel storage department for his steel sheets and an order placed 
with the punch press department for covers was all that was re- 
quired to keep him supplied with materials. Similarly, the shipping 
department had only to be notified once a week or so as to the prod- 
uct that was coming through for a particular purchaser. Further- 
more, long runs of a few simple products made for infrequent 
machine breakdowns and consequently few calls for help from the 
maintenance department. 

A drastic change in the market situation brought about a sweep- 
ing change in the activities and interactions of Joe Walker. Now, to 
keep the plant running, the salesmen were seeking and accepting 
orders for shorter runs and for a much wider varietv of barrels than 
in the earlier period. The short runs and the frequent changes in 
production and production lines called for Walker to interact much 
more frequently with foremen in the departments supplying him 
and with the foremen in the shipping department. In other words, 
he was thrown into much more frequent contact with indi\'iduals 
with whom he had never got along anyway. The change also put 
pressure upon his w^eakness in planning, for now at last there were 
complex activities to plan for the department. Finally, it seems that 
machines that are left alone break down much less frequently than 
machines that are constantly being adjusted for different types of 
production operations. The change therefore required ^^^alke^ to 
call for help much more often from the maintenance department, 
which was again a point of previously established frictions. 

266 • WiUiam Foote Whyte 

Elsewhere * I am describing this case in much more detail, in- 
cluding other changes than those which occurred in the work flow 
and production process, but perhaps even this highly simplified ver- 
sion suggests the importance of the impact of these factors upon 
organizational behavior. 

We can, of course, say that Joe Walker's orientation to the situa- 
tion changed in the course of these two years, but why concentrate 
on these elusive data when we can use data almost as concrete as 
the steel barrels that gave Walker so much trouble? Furthermore, 
it is only through examining such objective changes as I have de- 
scribed that we can explain the change in the situation and the 
change in Joe Walker's orientation to it. 

Other omissions by Parsons could be noted, but they all add up 
to the same point: by failing to deal in any adequate way with data 
that are abundantly observable and measurable, Parsons has chosen 
to erect his theoretical scheme on quicksand. 

These remarks regarding omissions will come as no surprise to 
Talcott Parsons. I made essentially the same points in a seminar 
discussion held with him at Cornell. At the time, if I remember 
correctly, he acknowledged that (1) the factors I described were 
indeed important, and that (2) he did not deal with them. 

This statement reflected the generous view of the research of 
others for which Parsons is justly known. Perhaps I should be 
equally generous in my reply and say in effect: "You go your way 
and I will go mine. There is room in the field of organization theory 
for both of us." However, I do not feel that this kind of tolerance 
is appropriate for the advancement of knowledge. I think we must 
agree that, from the standpoint of building organization theory, 
factor X is either important or unimportant— to simplify the argu- 
ment, let us ignore the gradations in between. If factor X is un- 
important, then it can safely be disregarded. If factor X is indeed 
important, then it cannot safely be disregarded. It does not make 
sense to say, in the same breath: "For the purposes of building or- 
ganization theory, factor X is important, but I choose to exclude 
factor X from my theory of organization." 

* William F. Whyte, Men at Work ( Homewood, 111., Irwin-Dorsey, 



Parsons is attempting to build a systematic theory of so- 
ciety, and therefore his system should apply to organizational be- 
havior. I have already indicated that I find this effort unsuccessful. 
I do not find here an acceptable theory of organization— which 
would certainly be too much to ask, given the scant attention Par- 
sons has devoted to organizations. But neither do I find the bases 
upon which a sound theory of organization might be built— and 
perhaps I might be permitted to ask for that much. 

This does not mean that I regard the work of Talcott Parsons as 
of no consequence. Although I now and then ask myself whether 
the intense effort necessary to penetrate the writings of Parsons 
can justify the value of the ideas to be found therein, I do feel that 
he has presented us with a number of very useful ideas. 

Others can review this sort of contribution much better than I, 
but let me mention several points so as not to leave the impression 
that I am simply being polite in my conclusion. 

It seems to me that the pattern variables are a most stimulating 
contribution. Whoever seeks to deal with the characteristic ways 
that people view the world around them will find that he gets useful 
leverage on his problems from the Parsons formulations. I find this 
approach particularly pertinent in dealing with intercultural com- 

As Henry Landsberger has pointed out, Parsons' analysis of the 
exchange of values that takes place between employer and em- 
ployee in an organization provides an approach whereby the think- 
ing of economists and sociologists can be more effectively brought 

Others may find any one of a number of other points equally 
stimulating or even more valuable. This suggests that the contribu- 
tion of Talcott Parsons will eventually not be in the area of system- 
atic theory building but rather in the creation of a number of 
provocative ideas which can be used by many students, whatever 
their theoretical orientations. 



Max Black 







A thorough investigation of Parsons' methodology 
would require consideration of the following questions: 

(1) What is to be understood by a "general theory" of 
action? (Are there analogues in the natural sciences? What 
are the advantages to be expected from such a theory? Is 
such a theory necessary or desirable in the social sciences? ) 

(2) How are the basic categories of Parsons' theory ob- 
tained? (Do they arise from previous empirical research? 
If so, how? Are there any acceptable tests of their validity? 
If so, what are they? What would be suflScient grounds for 
revising Parsons' conceptual scheme? ) 

(3) In what ways are the "pattern variables" related to 
the basic conceptual scheme? (Does that scheme require 


tliose variables? Are the five variables "exhaustive" of all possibil- 
ities? Would alternative modes of classifying "pattern variables" be 
compatible with the general intentions of the supporting theory?) 

(4) What kinds of definitions are provided for the basic terms 
of the theory? ( Are the terms univocal or "schematic"? How directly 
are they linked with possible observations? Are explicit definitions 
of the separate terms conceivable— or does the system stand as a 
whole in some kind of confirmable relation to experience?) 

( 5 ) How is Parsons' theory to be construed? ( Is it to be regarded 
as a set of highly general hypotheses about persons and social sys- 
tems? Or is it, perhaps, better viewed as a "frame of reference," 
whatever we take that to be? Is it, perhaps, best regarded as a 
terminology for the expression of substantive social theory?) 

I oflFer these examples as representative of the kinds of questions 
that need to be asked about Parsons' work, though I have neither 
the time nor the capacity to answer all of them. It will be noticed 
that the questions are concerned with problems of method; I am 
quite unable to judge the merits of Parsons' specific contributions 
to the sociology of the professions, the study of small groups, or to 
any other branch of the social sciences. 

My remarks will be almost entirely limited to the discussions 
contained in Toward a General Theorij of Action and especially to 
the "General Statement" and the monograph entitled "Values, Mo- 
tives, and Systems of Action" to be found in that work. (I speak 
throughout of "Parsons," though the works in question are in fact 
attributed to Parsons and Shils. ) 

II. parsons' conception of the purpose 


The theoretical constructions to be discussed have arisen 
from an attempt to unify the foundations of psychology, sociology, 
and anthropology— or, as we might put it, to provide a charter for 
the united social sciences. 

According to Parsons ( GTA 3 ) , the desired "general theory" will 
consist of "generalized hypotheses" together with a "formulation of 

270 • Max Black 

certain fundamental categories." The hypotheses are to permit "sys- 
tematic reformulation of existing facts and insights," or "codification." 
By displaying relations between hitherto disparate facts and gen- 
eralizations, they will point the way to further observations. The 
categories "enter into the formulation of this general theory." Hav- 
ing been selected in an orderly or "systematic" way (according to 
procedures to be examined later in this paper), they facilitate con- 
struction of an internally organized or "systematic" theory, rather 
than a mere aggregate of disconnected generalizations. In this way 
they help us to become "more aware of the interconnections among 
items of existing knowledge which are now available in a scattered, 
fragmentary form." (GTA 3) 

' Comments 

(1) Parsons is clearly right in distinguishing between the task 
of framing highly general hypotheses, derived from existing special 
hypotheses, and the task of devising an adequate terminology for 
their formulation. It is one thing to make broad statements, true or 
false, and another to propose a set of words for making such 
statements. The statements might be true, though expressed in an 
obscure terminology; or the words might have clear meanings, yet 
the statements be false or tautologous. 

(2) The two tasks are clearly connected: we have to judge the 
proposed terminology in the light of the statements it helps us 
to formulate. For instance, it would be impossible to have a set 
of statements simultaneously applying to anthropological, economic, 
and psychological data, unless we were provided with a vocabulary 
simultaneously applicable to all of these domains. 

(3) Parsons does not provide a list of his "general hypotheses": 
we are left to perform this crucial task ourselves, by collating his 
scattered remarks (cf. Section III infra). 

(4) Stress upon the need for "system" in presenting "hypotheses" 
and "categories" is highly characteristic of Parsons' thought. "Sys- 
tem" in this sense must not be confused with the sense in which 
there are, according to Parsons, systems of action. Parsons' theory 
is intended to be systematic in the dictionary sense of "arranged, 
or conducted, according to a plan or organized method." More 
specifically, Parsons' theoretical ideal demands a terminology con^ 


trolled by explicit principles of classification that generate a com- 
plete inventory of logically possible combinations. It is, as I see it, 
the ideal of an intelligible and exhaustive schedule. Given such 
a schedule, we may expect the associated hypotheses to exhibit a 
similar kind of order— that is to say, to result from a methodical 
scrutiny of a complete set of logical possibilities.* 

(5) While there is no objection of principle to the proposed way 
of trying to organize scientific disciplines into a "system," it is 
worth remembering that there are other and equally adequate 
modes of unification. Physics, chemistry, and biology are rapidly 
becoming "unified," so that the divisions between them appear 
increasingly to be based upon practical convenience in the division 
of scientific labor. But this has not come about by the imposition 
on all three of a single schedule of classificatory concepts. f 

III. parsons' basic assumptions 

I shall try to state in this section some of the "general 
hypotheses" whose truth Parsons takes to be sufiiciently established 
—the general truths about individual and social behavior that he 
feels called upon to recognize in his system. As I said above, these 
are not presented in any orderly fashion in Parsons' writings, and 
have to be extracted from discussion about related topics. I cannot 
expect to make the list complete. 

* Cf. GTA 20, f.n. 27. My account differs markedly from that given 
elsewhere by Parsons (GTA 49). His own criteria of "system" are 
(a) the "generality and complexity" of the theory, (b) the degree to 
which the various assertions of the theory are in exphcit deductive 
connection with one another, and (c) "the level of systematization; 
that is, . . . how far the theory is advanced toward the ultimate goals 
of science." Of these, the first and last are too vague to be of any 
service, and the second does not fit Parsons' practice. To anybody 
famihar with deductive procedures in mathematics and the natural 
sciences. Parsons' claim to "carry deductive procedures further than 
is common in the social sciences" (GTA 49) will seem surprising. 
There is very little strict deduction in Parsons' exposition, 
f For a recent discussion of some of the theoretical issues involved, 
see Paul Oppenheim and Hilary Putnam, "Unity of Science as a 
Working Hypothesis," Minnesota Studies in the Philosophy of Science, 
2:3-36 (1958). 

272 • Max Black 

(1) All human action is directed toward goals. 
Comment: I take Parsons' use of the word "orientation" to be a 

way of talking about goals. To be oriented toward G, is to have G 
as a provisional or terminal goal, and vice versa. It is important | 
to bear in mind throughout that the "goal" is the desired terminus 
as it appears to the actor. It is the end in view, rather than any 
terminus that may be imputed by an external observer. This impor- 
tant contrast is somewhat blurred by Parsons' admission of "uncon- 
scious" goals. 

(2) All human action is relational, in the sense of being a func- 
tion of the actor's innate needs (or "viscerogenic needs"), his 
acquired orientations, and the particular situation in which he 
finds himself. 

Comment: Here, as before, it is important to stress that the 
"situation" in question is the situation as it presents itself to the 
actor himself— the "subjective" or "psychological" situation, as it 
were. The "orientations" can be thought of as acquired predisposi- 
tions to respond in certain ways to given stimuli. It is central to 
Parsons' approach to insist that these predispositions are largely 
the products of the goals and standards of the social system to 
which the individual belongs. 

(3) All human response to stimuli has two distinct dimensions- 
is simultaneously cognitive and cathectic. 

Comment: I say "response" here, rather than "action," because 
the distinction, so far as I can see, is intended to apply to covert 
changes in the actor as well as to overt reactions to the demands 
of a situation. Parsons sometimes refers to both of these as "action." 
Pages could be written about the meanings to be imputed to 
Parsons' key terms, "cognitive" and "cathectic." At the crudest level 
of common sense I would try to translate "to cognize" as "to per- 
ceive, believe, to think, in short to do anything with respect to 
which questions of truth or falsity may arise"; similarly, "to cathect" 
might be rendered as "to be attracted or repelled by, to like or 
dislike, to want or not to want, in short to do anything with respect 
to which questions of personal satisfaction or dissatisfaction can 
arise." These formulas are vague, to be sure, but no more so than 
Parsons' own accounts. Thus Parsons variously describes "cathexis" 
as an "attachment" (GTA 5), a "response" (GTA 8), a "state" 


(GTA 10), and so on. These variant formulations are not easily 
reconcilable with one another. 

(4) All human action involves selection between alternative 
orientations and responses. 

Comment: The process o£ deciding between alternatives is called 
evaluation (GTA 11) 

(5) Selection (or evaluation) involves the use of standards. 
Comment: Standards, also called "norms," may perhaps be 

thought of as rules or prescriptions for making choices. I pass over 
various distinctions that might be made. It is worth pointing out 
that Parsons repeatedly thinks of "evaluation" as a problem of 
"allocation" of scarce resources among conflicting demands and 
interests. Such allusions to analogues in economics are fairly com- 
mon in his writing. 

(6) All interaction between actors involves complementarity of 
expectation, in the sense "that the action of each is oriented to the 
expectations of the other." 

Comment: Expectation falls on the side of the "cognitive": one 
might wonder why there should not be also a parallel "cathectic" 
complementarity, with, as it were, the desires of the self "oriented" 
to the desires of the other. I do not find this in Parsons— perhaps 
because it is also absent from Mead, who is the acknowledged 
progenitor of this segment of Parsons' theory. 

(7) Orientations and actions are organized in systems. 
Comment: Here, "systems" must be construed as analogous to 

organisms, in the biological sense. *^ They are conceived to have the 
crucial property of being "boundary-maintaining" and "structure- 
maintaining"— they resist external attack, and exercise an internal 
control over their components analogous to "homeostasis." 

(8) All the above principles apply to social systems of all levels 
of complexity, up to and including the total society, as well as to 

Comment: This is so broad in its implications that it can hardly 
be treated as coordinate with the seven statements that precede it. 
However, it is so integral to Parsons' thought that I have felt its 

* Cf. Talcott Parsons, "Some Comments on the State of the General 
Theory of Action," American Sociological Review 18:623 (1953). 

274 • Max Black 

inclusion to be necessary. It is characteristic of Parsons' procedure 
to apply the same kind of language hoth to individuals and also 
to groups, professions, and so on. 

General Comments on Parsons' Assumptions 

Given the eight propositions listed above, we have enough 
illustrative material to raise the crucial question of method. I take 
this to be: Are Parsons' "assumptions" properly to be regarded as 
empirical generalizations? 

Some merits of Parsons' approach. Since many of my subsequent 
comments will be critical or skeptical, I think it proper to begin 
by registering admiration for the reach of Parsons' theoretical 
scheme. The great achievements of scientific theory have not been 
made by timid men, anxious to stay close to the facts at hand, or 
to what their contemporaries took to be such "facts." Science needs 
men of imagination, willing to incur the risks of speculative con- 
struction. The need is particularly urgent in the social sciences, 
where the temptation is great to waste time in questionnaire con- 
struction, the factor analysis of trivial data, and other ways of add- 
ing to the bulging files of unread "research." Only those who have 
faced the exacting problems of theory construction are entitled to 
throw stones at Parsons' stimulating and pioneering labors. It has 
the great merit of being a theory, an honest and ingenious attempt 
to provide a basis for linking together and understanding lower- 
level generalizations. Right or wrong. Parsons' system raises impor- 
tant questions, suggests new ideas, and provides unexpected leads 
for empirical investigations. 

Any reader must be impressed also by the complex architecttire 
of the system. The intricate connection of concepts typical of 
Parsons' approach, though it adds to the diflficulty of understanding, 
also promises fruitful application. So much by way of preliminary 
appreciation and tribute. 

7s the theory basically "static"? A curious feature of Parsons' 
theory of action, remarked upon by several critics,* is the extent 

* E.g., by G. E. Swanson, in his article, "The Approach to a General 
Theory of Action by Parsons and Shils," American Sociological Review 


to which action, in any ordinary sense of that term, fails to get 
discussed at all. As we have seen, the theory leans heavily upon 
the notion of orientation or, as we might prefer to say, attitude. 
Now an orientation or attitude has to be construed as a state of a 
person or a social system: it is an abstraction from the condition 
of a given "actor" in a given social field or social situation at a 
given moment. (The very word "orientation" suggests something 
static— the direction in which a body is pointing.) Certainly, an 
attitude must also be regarded as a predisposition to act in a 
certain way, so that the operational tests of a given actor's having 
a given attitude will have to consist of observations of his acts, 
coupled with inferences as to how he would act in other situations. 
But if reference to actions in the ordinary sense thus enters into 
the very definition of an attitude, it still remains true that the 
end-product is a description of how the actor stands ( what internal 
and external stresses determine his position of momentary equilib- 
rium). There is consequently a serious question whether any 
description of this type, valuable as it may be in other respects, 
will yield predictions as to how the actor will movte (i.e., will act 
in the ordinary sense of the term). We may well remind ourselves 
at this point that what is called the "statics" of ordinary material 
bodies (the theory of bodies under equilibrium) has to be sup- 
plemented by independent mechanical principles before we are 
in a position to say anything about the motions of material bodies. 
Statics can be regarded as a particular case of dynamics, but not 
vice versa. And this is so, in spite of the fact that definitions of the 
"forces" acting upon a material body in equilibrium involve refer- 
ence to the ways in which the bodies in question would begin 
to move in certain test conditions. 

Parsons anticipates this criticism in a section of GTA entitled 
"Descriptive and Dynamic Analysis." (GTA 6) His reply amounts 
to saying that the very same variables will be needed in both the 
static and the dynamical theories, and that it is uneconomical to 
formulate dynamical questions before the questions of statics have 
been answered. This may be readily conceded. But he apparently 
overlooks the need for new variables and new principles when we 
introduce dynamical considerations. It seems to me that his choice 
of concepts restricts him to discussing cases of equilibrium or quasi- 

276 • Max Black 

equilibrium (cases in which changes are so small or so gradual as 
to be negligible) and in this way limits the applicabihty of his 

I think we can generalize somewhat and say that Parsons, con- 
sciously or not, is primarily interested in the equilibrium conditions 
of social system. The following is a characteristic statement: "The 
most general and fundamental property of a system is the inter- 
dependence of parts or variables . . . [which] is order in the 
relationships among the components which enter into a system. 
This order must have a tendency to self -maintenance, which is very 
generally expressed in the concept of equilibrium." * 

In view of the influence upon Parsons' thought of economic 
models, we may remind ourselves of the notorious difficulties in 
classical economic theory of handling problems of change and 
development, f 

I conclude that in the present stage of development of Parsons' 
theory, it should be regarded primarily as a theory about "social 
statics," to be appraised by criteria appropriate to such an under- 
taking. J In particular, we should not expect Parsons to be able 

* There follows the revealing footnote: "That is, if the system is 
to be permanent enough to he worth study, there must be a tendency 
to maintenance of order except under exceptional circumstances." 
(GTA 107; italics added.) 

f Cf. the following statement: "I find myself in the curious position 
of, on the one hand, saying that economics has very important con- 
tributions to make to the study of social dynamics, and, on the other 
hand, finding it almost necessary to deny that there can be any such 
thing as economic dynamics." Kenneth E. Boulding, "Economics as 
a Social Science," The Social Sciences at Mid-Century (Minneapolis: 
University of Minnesota Press, 1952), p. 82. Boulding looks hopefully 
to the sociologist to "take into account some of the more important 
variables of actual social dynamics" {ibid.) and in this way to invest 
economics with genuine predictive power. 

\ I will add one small but perhaps significant piece of evidence. In 
citing examples of the use of his scheme "as a direct instrument of 
new empirical research," Parsons mentions an attempt "to use and 
develop this type of theory in predicting from the characteristics of 
families, peer groups, and schools, and the place of a boy in them, 
what place in the occupational system he will come to occupy." 
American Sociological Review, 18:630 (1953). This type of predic- 
tion, from one state of quasi-equilibrium to another, is strikingly 
reminiscent of the so-called method of "comparative statics" in 


to predict how social systems will change, though we might rea- 
sonably expect him to tell us which conditions of such systems 
might be expected to be relatively stable. 

Is the theory iion-empirical? A number of Parsons' readers are 
likely to be troubled by the apparent distance of his basic concepts 
from direct observations. Certainly, no superficial inspection of the 
world will yield "attitudes," "subjective goals," "internalized stand- 
ards," and the other conceptual instruments in Parsons' armory. 
So an uneasy suspicion may arise that Parsons has provided a free- 
floating linguistic system, capable of gratifying those who have 
succumbed to its formal charm, but resisting any prosaic mooring to 
observational criteria. 

I believe such fears are unwarranted. The concepts instanced are, 
of course, "high-level" ones, "constructs" unamenable to explicit 
definition in terms of observables. But the time should long be past 
when this was regarded as a defect. Physics, "the science that has 
made good," contains any number of concepts that are similarly 
remote from experience. So long as the system in which the con- 
cepts are embedded is furnished with an adequate supply of "coordi- 
nating definitions" for drawing observational consequences from 
theoretical premises, that is all that can reasonably be expected. 
I see no reason why this should in principle be impossible in 
Parsons' case. The ingenuity of the experimentalist can be trusted 
to invent reliable tests for the presence of any of the concepts that 
Parsons needs, though I do not wish to underrate the difiiculties 
of this program in particular cases. Parsons' theory is a good deal 
"closer to immediate reality" than, say, quantum physics, and should 
not be judged more harshly than the latter on these grounds. 

On the other hand, I am inclined to wonder whether the type of 
theory that Parsons is presenting can reasonably be expected to 
be causal in form. It will be remembered that the "high-level" or 
"molar" laws of physics take the form of general principles restrict- 
ing admissible types of transformation or change (as in the case 
of principles of "conservation"). While we cannot exclude the pos- 
sibility of causal connections, in some sense of that expression, 
between "emergent properties," their instrumental function in pro- 
viding deductive connections between the components of a theo- 
retical system recommends caution. To state my difficulty a httle 

278 • Max Black 

more definitely: I would be surprised if genuinely causal laws 
could be shown to hold between "orientations," "goals," and the 
like. It seems to me, as it were, a priori likely that we must look 
for the causal chains in the discernible pressures upon, and resist- 
ances within, individuals. If this is so, Parsonian principles will 
have to be regarded as laying down general restrictions upon the 
forms such changes can take— i.e., as specifying a framework for 
possible laws, rather than as themselves constituting the laws we 
ultimately hope to find. 

Where do the concepts come from? Parsons does not tell us how 
he obtains the basic concepts of his theory (though he does say 
a good deal about the influences that have guided his own thinking, 
mentioning J. L. Henderson, Max Weber. Freud, Hull, and many 
others). Now, if we had a clear view of the process by which the 
concepts are obtained, we might be in a position to judge whether 
they have been well chosen and in what ways the choices might 
be improved. (I shall not here consider the pragmatic test of how 
well the system works in stimulating fruitful experiment and 
observation— time alone can settle that.) 

The point has some importance in view of the impression I get 
of the arbitrariness of some of the decisive choices made by Parsons. 
To take a single example of great importance for appraising his 
scheme: What is the justification for the "cognitive-cathectic" con- 
trast that runs through the whole of the theory? I have already 
hinted at the shifts in meaning that the term "cathectic" manifests 
in Parsons' writings. The cognitive-cathectic contrast is hardly 
more than the layman's crude contrast between "thinking" (with 
believing, perceiving, etc., thrown in) and "feeling." It seems un- 
likely that genuine science is to be expected in terms of distinctions 
as crude and unsystematic as these. I can illustrate what I mean 
from a field in which I feel a good deal more at home— the theory 
of language. For a time there was current here a contrast between 
"referential" and "emotive" discourse that closely parallels Parsons' 
distinction between the "cognitive" and the "cathectic." It has 
become increasingly plain, however, that the multifarious uses of 
language are not to be understood in terms of so crude an opposi- 
tion, a more detailed and flexible system of organizing concepts 
being needed to do justice to the complexity of linguistic phenom- 


ena. And I am inclined to think that the same must be true of 
Parsons' scheme. I beheve an over-simple psychological analysis 
pervades Parsons' thought and thereby limits its usefulness.* 

On the whole, it seems to me, the component concepts of Parsons' 
scheme are laymen's concepts in the thin disguise of a technical- 
sounding terminology. 

The following might be the result of trying to express Parsons' 
postulates in plain English: 

( 1 ) "Whenever you do anything, you're trying to get something 

(2) "What you do depends upon what you want, how you look 
at things, and the position you find yourself in." 

(3) "You can't do anything without thinking and having feelings 
at the same time." 

(4) "Human life is one long set of choices." 

(5) "Choosing means taking what seems best for you or what 
others say is the right thing." 

(6) "When you deal with other people, you always have to 
take account of what they expect you to do." 

(7) "There's a lasting pattern to the way people behave." 

(8) "Families, business firms, and other groups of persons often 
behave surprisingly like persons." 

I think these aphorisms contain nearly all of the content of the 
Parsonian principles I listed in Section III above. 

Perhaps this shows how close Parsons has remained to the wisdom 
of the hoi polloi. But one may wonder whether it is plausible for 
fundamental social theory to be so close to common sense. If the 
history of the development of the natural sciences is any guide, 
fundamental social theory will have to employ recondite notions, 
at a considerable remove from direct observation, in order to have 
any hope of providing an adequate framework for research. As 


* The contrast between "thinking" and "feehng" is reminiscent of the 
ancient tripartite division of the faculties. But what has happened to 
the "conative" in Parsons? The functions of the will in the older 
psychology seem to have become absorbed in the "allocative" func- 
tions of Parsons' "evaluation"— as if the will were a kind of referee 
between thought and sentiment. I would like to see a competent 
psychologist engage in a detailed criticism of Parsons' psychological 

280 • Max Black 

Ernest Nagel has said, the concepts of a comprehensive social theory 
"will have to be apparently remote from the familiar and obvious 
traits found in any one society; its articulation will involve the use 
of novel algorithmic techniques; and its application to concrete 
materials will require special training of high order." * Parsons' 
theory does not meet these requirements. 

Two types of ambiguity in Parsons. I have been suggesting that 
Parsons' principles are close to the level of proverbial wisdom. 
Now it is characteristic of proverbs, and one reason for their use 
as a substitute for precise thought, that they embody ambiguities 
making them indefinitely adaptable to almost any circumstances. 
("Look before you leap!" Of course. But then it all depends on 
what you recognize as the "leap" and what is to count as "looking." ) 
I want to argue that Parsons' principles manifest the same peculiarity. 

Consider the principle that all human action is directed toward 
goals (the first principle listed in section III above). We might 
narrowly construe "goals" to mean something like "explicit end-in- 
view." In that case, the principle is plainly and obviously false. 
If I am trying to hammer a nail into a hole, I certainly do have an 
explicit end-in-view, so that my action can be properly described 
as directed toward a goal, in the narrow sense of that term. On the 
other hand, if I am smoking a cigarette, I have no explicit end-in- 
view, neither of consuming the cigarette, nor of soothing my nerves, 
nor of finding something to do, nor anything else. In the sense of 
goal in question, nearly all human action is not directed toward 
a goal. But, on the other hand, if we allow "goal" to be construed 
suflficiently widely for "unconscious goals" f to be admitted, the 

* "Problems of Concept and Theory Formation in the Social Sciences," 
Science, Language and Human Rights (Philadelphia: American Philo- 
sophical Association, 1952), p. 63. 

f "Another source of complexity and possible misunderstanding is the 
question of whether ego's orientation to an object must, in whole or 
part, be conscious. The answer is quite clear; it is not necessary. The 
criterion is whether ego acts toward the object in a meaningful way 
so that it is reasonable to interpret his action as based in his orienta- 
tion as to what the object is, has been, or is expected to be." 
(GTA 104) Similar references to unconscious factors abound in 
Parsons' work. Thus he speaks of "evaluation, which operates uncon- 
sciously, as well as deliberately." (GTA 14; italics added.) Each of 
the factors mentioned by Parsons has to be understood as sometimes 
operating unconsciously. 


matter looks different. At once we find a strong inclination to say 
that every action must be "goal-oriented," because we are now 
determined to see every action as if it had an explicit and conscious 
goal. The "principle" now runs the risk of degenerating into a 
sterile tautology, resulting from a stretching of the key terms 
( "action," "goal," etc. ) to a point at which they apply to everything 
and mean nothing. I suspect an endemic ambiguity of scope in 
Parsons' use of his key terms, a constant vacillation between narrow 
and broad senses. 

A different kind of ambiguity springs from the intention to 
apply a single system of concepts simultaneously to social systems 
as well as to persons (see principle 8, Section III, page 273 supra). 
That there are striking analogies between social systems and per- 
sons may be freely conceded; yet it is equally obvious that concepts 
literally applicable to persons cannot be transferred to groups or 
other social systems without systematic alterations in their mean- 
ings. An organization cannot literally have a "goal" in the sense 
in which a person has one; it cannot "cathect" (have feelings for 
or against) in the same sense in which its officers do; if it "internal- 
izes" standards of the enveloping culture, the way that this comes 
about must be substantially different from that in which a single 
man is taught to accept and conform to the standards of the groups 
to which he belongs. These are methodological platitudes I would 
hesitate to repeat were it not that Parsons shows so little awareness 
of them. I am disturbed by the systematic ambiguity that pervades 
Parsons' discussions, and the freedom with which he passes from 
considerations, plausible when made about persons, to assertions 
having the same verbal form, but necessarily having a different 
meaning when applied to social systems. 

Are the principles empirical generalizations? I have already 
argued that by taking the key terms of Parsons' scheme in "stretched" 
senses, it is possible to convert them into analytic statements, whose 
truth is guaranteed by the implicit definitions of their component 
terms. I want now to consider what the situation \^dll be if we do 
not succumb to the temptations of this sleight-of-hand, but insist 
upon attaching relatively precise and narrow senses to "goals," 
"orientations," and the like. Parsons' principles can then no longer 
be regarded as universally applicable; but can they not be still 

282 • Max Black 

regarded as empiricar generalizations having very wide application?* 
Could we not say something like the following: 0£ a very large 
class of human actions it is in fact true that they are directed toward 
a goal or end-in-view? To make this interpretation work, it would, 
of course be essential so to define the "class of human actions" in 
question that being "directed toward a goal" should not be part 
of the definition of that class; but it does not seem impossible to 
do this. It seems plausible, accordingly, to hold that the principles, 
understood in the way sketched above, are true empirical general- 
izations of wide scope, and that accordingly. Parsons is right in 
his claim to have rooted his theory in solid fact. 

I want to argue, however, that this is a mistake, though a plausible 
one. Consider what the world would be like if the alleged "empiri- 
cal generalizations" were false. This requires imagining a world in 
which the classes of action in which men do have ends-in-view— 
sharpening pencils, starting cars, attaching stamps to envelopes, 
and so on— occur, without any ends-in-view. The more we try to 
imagine such a "world" the more inconceivable it becomes. It 
would have to be a world in which something like present human 
activity occurred, but in which any reference to "what I am trying 
to accomplish" would be utterly pointless. A world in which nobody 
would ever be trying to do anything, or to achieve any objective— 
a world of aimless activity. Perhaps such a description might fit 
an army of robots; but it would certainly be grotesque to apply 
it to anything that we would want to call human activity. I am 
arguing that close examination will show the concept of "purposive 
activity" to be a component in our concept of "human action." If 
this is so, the conception of the principles in question as being 
empirical generalizations of wide scope is an illusion. That human 
beings often have goals or ends-in-view is not a fact, but rather 
something that follows from our conceptions of what it is to be 

* I find another writer speaking of "two strategic generalizations" 
which he formulates as follows: "(a) Some human behavior is 
normatively oriented, being shaped by values, symbols, signs, etc.; 
(b) most human beings Hve in groups at any given time in their hfe 
cycles, and their behaviour is influenced by their interactions with 
others." Alvin W. Gouldner, Sociology in the United States (Paris: 
UNESCO, 1956), p. 35. I doubt whether these statements should be 
regarded as expressing generahzations. 


human. And much the same can be said about Parsons' other 

Summanj conclusions. I am forced to conclude that Parsons' 
principles are not founded in empirical generalizations, in any 
plausible sense of "empirical generalizations." He has provided us 
with a web of concepts, whose correspondence with the concepts 
which laymen use for thinking about social relations and human 
action is barely disguised by a new terminology. 

If this verdict should be sustained, some disturbing consequences 
would result for Parsons' claim to have provided a scientific frame- 
work for the social sciences. For it is easy enough to provide some 
set of definitions or some conceptual scheme: the diflBculty is to 
provide one that is not capricious or arbitrary. The su:preme virtue 
of a scientific classification, whether in physics, chemistry, or biol- 
ogy, is that it arises from, and is in some sense demanded by, a 
system of well established empirical generalizations and theories. 
Mendeleef's classification of the elements, to take a familiar exam- 
ple, arises from a wealth of empirically established regularities 
concerning the properties of chemical substances and compounds. 
This accounts for its superiority over the earlier classifications of 
the alchemists. Similar claims can be made on behalf of the 
superiority of biological classifications into species, genera, and 
so on, over the crude classifications of common sense: Only with 
the gradual discovery of the laws of heredity did it become possible 
to establish a truly scientific way of describing animals and plants. 
If the parallel should hold for the social sciences, we would have 
to say that the elaboration of a conceptual scheme, such as Parsons 
oflFers us, would have to await a wealth of well-founded empirical 
generalizations. But I have argued that this is not the way Parsons 
proceeds. Perhaps the reason is that the requisite generalizations 
are not yet available. But there are no short cuts to a scientific 
classification of human actions. 

IV. parsons' conception of the "pattern variables" 

The general theoretical framework I have so far been dis- 
cussing resulted from an attempt by specialists in various branches 

284 • Max Black 

of the social sciences "to find the greatest possible measure of 
common ground." * The weaknesses emphasized above may perhaps 
be attributed to this provenance: it is not rare to find attempts for 
the greatest common measure of agreement, in science as in politics, 
relying upon a lowest common denominator of significance. To 
find a more distinctive and, potentially, more fruitful contribution, 
we may turn to Parsons' exposition of the so-called "pattern 
variables." f 

It is gratifying to find, for once, a formal definition of a key 
term in Parsons' system: "a pattern variable is a dichotomy, one 
side of which must be chosen by an actor before the meaning of 
a situation is determinate for him, and thus before he can act with 
respect to that situation." (GTA 77) Parsons also likes to speak of 
"dilemmas of choice." {Am. Soc. Rev. 1953, p. 622). 

We gather, then, that a pattern variable" is a set of two mutually 
exclusive alternatives. But alternative whats? Within the space of 
two pages we are told that the pattern variables are "characteristics 
of value standards," (GTA 78) "can be used to characterize differ- 
ences of empirical structure," (GTA 79) are "categories" (ibid.) 
and "inherently patterns of cultural value-orientation" {ibid.; italics 
added in each quotation). Whatever else they may be, these 
elusive "pattern variables" have an amazing power to be different 
things on different occasions. I think it would not be unfair to 
call them "chameleon concepts." For some of the ambiguities infect- 
ing them are certainly intentional: on the same page, we find one 
of the alternatives of the first "pattern variable" called a "normative 
pattern" (in its "cultural aspect"), a "need-disposition" (in its 
"personality aspect" ) , and a "role-expectation" ( in its "social system 
aspect"). Here we have an extreme instance of the "systematic 
ambiguity" to which I drew attention above. 

At the risk of ignoring a great deal of what Parsons himself 
regards as of crucial importance for his system, I shall now confine 

* Parsons, "Some Comments," p. 621. 

I "The core of the more personal contributions which Shils and I have 
made is to be found, in our opinion, in what we have called the 
'pattern variables.' " Parsons, op. cit., p. 622. 


my remarks, to the meaning o£ the expression "pattern variable" 
when apphed to single persons. Here, we might take a pattern 
variable to be: 

a set of two mutually exclusive kinds of choice that face any 
given individual prior to action. 

A secondary meaning, that is sometimes appropriate for under- 
standing Parsons, is: 

a set of two expressions standing for the mutually exclusive 
kinds of choice just mentioned. 

We can now tabulate the types of choice presented by Parsons 
as follows: 



1. to get immediate gratification to exercise self-restraint in the light 

of long-term considerations 

( "aff ectivity— aflFective neutrality" ) 


2. to serve self-interest to serve the interest of a group to 

which one belongs 
"self -orientation— collective-orientation" 


3. to treat an object or another per- to take account of the particular re- 

son as falHng under some general lations in which the object or person 

principle in which there is no stands in relation to oneself 

reference to oneself 

"transcendence— immanence" 

286 • Max Black 


to treat an object or another per- to treat it or him in the hght of 

son in the Hght of "what it is" what it or he may be expected to do 

(its supposed quaHties) 

"ascription— achievement" 


5. to respond to many aspects of to respond to some selection of those 

the object or person aspects. 

"diffuseness— specificity" 

In discussing the above scheme of classification, I shall be par- 
ticularly concerned to decide whether Parsons is right in his claim 
that the sets of alternatives offered constitute a "system covering 
all the fundamental alternatives which can arise directly out of the 
frame of reference for the theory of action." (GTA 88) But I shall 
not confine myself to this question alone. 

Miscellaneous Comments on the Pattern Variable Scheme 

1. The entire scheme rests upon the supposition of choices made 
by a given person in a social situation. Now, if "choice" were 
understood in a narrow and emphatic sense, it would be patently 
false to say that everybody has to make five choices of the kinds 
sketched above. Only very rarely is it the case, for instance, that 
anybody chooses to behave selfishly rather than altruistically (cf. 
the second pair of alternatives in the last section). But once we 
admit "choices explicit or implicit" (GTA 78, italics added) a dis- 
turbing element of the fictitious is allowed to enter at the ground 
floor. Here we have another striking instance of ambiguity in 
Parsons' thought. 

2. It would be hard to imagine more distressing choices of 
technical terms for labeling the distinctions invoked. Apart from 
being barbarous neologisms, and correspondingly hard to remem- 
ber, they have a pronounced tendency to suggest misleading or 


irrelevant associations. These severe practical inconveniences are 
exacerbated by Parsons' habit of using a variety of different descrip- 
tions to characterize what is intended to be a single item in the 
scheme of classification. For example, Parsons talks of the first horn 
of the first dilemma in terms of "impulse" and "gratification" and 
"permissiveness." Even if these are inseparable, they are not identi- 
cal. To use them indifi^erently as defining characteristics is to invite 
gratuitous confusion. 

3. An ancient principle of classification demands that a single 
intelligible principle of organization (a single "fundamentum divi- 
sionis") shall govern the scheme proposed. Now Parsons' scheme 
seems to use diflFerent types of principles at each step; the relation 
between the various principles remains obscure, in spite of numer- 
ous explanatory remarks by their author. 

4. Parsons insists that "the variables as we have stated them are 
dichotomies and not continua." ( GTA 91 ) I do not understand why 
he should regard this as important. Whatever the reasons for his 
contention, it seems plainly false in respect of some of the sets of 
alternatives. Thus, the fifth "dilemma" ("diflFuseness-specificity") is 
obviously a matter of degree, and is even presented as such: "how 
broadly is he (the actor) to allow himself to be involved with 
the object?" (GTA 83; itafics added). 

Are the pattern variables required by the general theory? Parsons 
repeatedly says that the pattern variables are required by the gen- 
eral theory: "the five pattern variables formulate five fundamental 
choices which must be made by an actor when he is confronted 
with a situation." (GTA 88) To this the following objections may 
be raised: (1) Given that the "evaluative mode of orientation" is 
regarded as a kind of controller of the other two, there should be 
three types of choices connected with these modes— that is to say, 
to evaluate or not to evaluate, and if the second, to prefer the 
cognitive orientation to the cathectic, or not. Parsons' objection that 
the "cognitive and cathectic modes of motivational orientation are 
so inseparable as to abnegate [obviate?] any problem of primacy" 
(GTA 88-9) seems to me dogmatic. Once it is recognized that all 
the "choices" to which he refers involve the more-or-less, there 
seems no reason for not also recognizing a "cognitive— cathectic" 

288 • Max Black 

When we turn to the "dilemmas" that concern attitudes toward 
the "object" (the last three of the five dilemmas) Parsons' pro- 
cedure appears even more arbitrary. For it is easy to think of any 
number of other ways of classifying the selected attitude of the 
"actor" to his objects. Why not introduce the "choice" between 
treating the object as a person or as a member of some social 
system? Or between taking account of or ignoring another person's 
"evaluative" aspects? Or, to instance an altogether different type 
of choice, between considering short-term factors and long-term 
ones? Some of these suggestions may seem pointless to anybody 
who wishes to control some empirical field of social research. It is 
not my purpose to offer them as serious alternatives, but merely to 
illustrate my contention that Parsons' own choices are not dictated 
by exigencies of logic and formal completeness, but at best corre- 
spond to what he regards as worth emphasizing within the web 
of interlocking concepts he has delineated.* 


I have now come to the end of what has been a laborious 
investigation. By directing attention to vulnerable aspects of Parsons' 
theory, I have run the risk of seeming to neglect the many provoca- 
tive remarks about special topics which make his prolific output 
of papers and books valuable. But I am sure he would himself 
wish to be judged by the contributions he has hoped to make to 
the integration of the social sciences. I have tried to show why 
I judged this attempt to have been less than successful. And I have 
not concealed my dismay at the conceptual confusions that in my 
judgment pervade the entire structure. Whether it would be pos- 
sible to introduce the requisite clarity into Parsons' system without 
altering its objectives or its main features, I seriously doubt. Indeed, 
I am inclined to wonder whether the social sciences are yet ripe for 
the kind of theory that Parsons and his associates have been seeking 
to construct. 

* It is interesting in this connection to notice that Parsons now 
recognizes six pattern variables. See Parsons, "Some Comments,' 
p. 624. 



Andrew Hacker 




In 1872 Karl Marx stood up before a public meeting at 
The Hague and uttered the following words: "We know 
that the institutions, manners and customs of the various 
countries must be considered, and we do not deny that 
there are countries, like England and America, . . . where 
the worker may attain his object by peaceful means." * It 
is remarks like this which turn scholarly heads gray. For 
in the space of several seconds Marx tore an all but fatal 
gash in the theory of history he had so painstakingly 
developed in his formal writings. The bourgeois state and 
society, Marx had insisted, had to be overturned by force 
and violence if the working class was to inaugurate an 
eflFective dictatorship as a prelude to the communist Utopia. 
Violent means were imperative if the values and institutions 
of capitalism were to be obliterated for aU time: the bour- 
geoisie would not mend its ways voluntarily and, unless 
destroyed, would bend every effort to sabotage the sociahst 
revolution. This, at least, is the substance of Marx's theory. 

* Quoted in Hans Kelsen, The Political Theory of Bolshevism ( Berke- 
ley and Los Angeles: University of California Press, 1948), p. 41. 

290 • Andrew Hacker 

The conscientious scholar, well versed in the theoretical literature 
of Marxism, might wish that Marx had never shown up at that meet- 
ing at The Hague: the remark about "peaceful means," despite the 
qualifications about national variations, simply does not fit into the 
general theory of history. Conscience, of course, precludes the con- 
cealing of uncomfortable evidence. Perhaps there were two Marxes 
—Marx the academic theorist and Marx the organizer of the First 
International. Perhaps, and perhaps not. At all events the scholar's 
position is a difiicult one— and it is not irrelevant to a consideration 
of Talcott Parsons. 

Parsons has no book entitled Polity and Society, and his brief 
remarks on politics in The Social System are clearly undeveloped. 
To gain an understanding of his political theory it is necessary 
to refer to his "Hague Speeches": occasional papers on a miscellany 
of political subjects. Four of these essays will be discussed here. 
All of them deal with questions of class, power, and politics as 
they relate to contemporary American society. Two profess to be 
special— that is, political— applications of the general system which 
is elaborated in his larger works, and all have the virtue of dealing 
with a specified society so that theoretical conclusions may be 
ranged against the available data. Insofar as Parsons' political anal- 
ysis is "derived"— a favorite word of Parsons'— from his formal 
system, an analysis of that analysis may throw some light on assump- 
tions which underlie the larger system. But the opening caveat is 
still in order: these are occasional essays and they were written for 
specific purposes. Students should think twice before using them 
as pebbles to derail the Twentieth Century Limited. It may well 
be that there are two Parsonses— the political and the sociological, 
and the two have yet to meet in a consistent way. This paper will 
attempt to show a number of junctures at which his politics and 
sociology are significantly relevant to each other. 

The "conservative" bias in Parsons' writings has been remarked 
upon by more than one commentator.* The central place he gives 

* See Lewis A. Coser, The Functions of Social Conflict ( Glencoe, 111. : 
Free Press, 1956), pp. 21-24; Barrington Moore, Political Power and 


to a theory of "equilibrium" in his system is made to be convincing 
evidence that the emphasis is on underlying social consensus rather 
than on continual, even irreconcilable, conflict. Insofar as the equi- 
librium idea is considered here, it will be with reference to Parsons* 
view of American politics, and not his social system as a whole. It 
should be pointed out right now that the epithet "conservative" 
is a deceptive one, and one undeserved by Parsons. While he 
shares some of the philosophical assumptions of a man like Edmund 
Burke— and these will be noted— he is on the whole a "liberal." This 
ideological commitment appears at two levels. On the more transi- 
tory plane Parsons' liberalism expresses itself in a partisan sense: 
in his approach to the proper functions of government he is sym- 
pathetic to a greater— but not overextended— assumption of public 
responsibilities for the general welfare. To be specific, he is one 
of the liberal-intellectuals of the Democratic Party, one of the 
Eggheads. In a more profound sense his liberalism is more his- 
torically based: it is the ideology of John Locke and John Stuart 
Mill, the ideology of political liberty and a free society. The two 
liberalisms, of course, go hand in hand, but it is best to keep them 
analytically separate in this discussion. It does not matter which 
label is attached to an individual's political thinking so long as we 
are aware of the substance of his ideas. It will, for purposes of 
convenience, be proposed that Parsons is a liberal: that his view 
of society is the conventional liberal one that has characterized 
academic thinking in the social sciences. 

In 1955 Parsons wrote an article for The Yale Review entitled 
"Social Strains in America," dealing with the problem of the attack 
on civil liberties which was then overt and widespread. Far from 
being a facile journalistic attack on the Wisconsin demagogue, it 
was a sophisticated analysis of tensions underlying recent American 
development. "McCarthyism can be understood as a relatively 
acute symptom of the strains which accompany a major change in 
the situation and structure of American society," he says. "The 
strains to which I refer derive primarily from conflicts between the 
demands imposed by the new situation and the inertia of those 

Social Theory (Cambridge: Harvard University Press, 1958), pp. 
122-25; C. Wright Mills, The Sociological Imagination (New York: 
Oxford University Press, 1959), pp. 44-49. 

292 • Andrew Hacker 

elements o£ our social structure which are most resistant to the 
necessary changes." * The new situation revolves largely about 
the fact that America has assumed global responsibilities which 
are both expensive and hazardous. By no means all Americans are 
as yet accustomed to this unsettled condition, and the fear of 
defeat at Soviet hands engenders anxieties at both conscious and 
unconscious levels. On the structural level there is the rapid growth 
of industrialization, but a growth without the stabilizing element 
of an antecedent feudal class structure. The result has been the 
emergence of an open society which, to put it simply, is too open. 
The class structure, such as it is, is based almost entirely on occupa- 
tional roles: this means that individuals find their expectations in 
life but weakly established, and their aspirations frequently frus- 
trated. In a more specific vein. Parsons points out that many busi- 
nessmen are angered about increasing government intervention in 
their hitherto private affairs. New men of economic power in the 
hinterland are jealous of the influence and prestige still possessed 
by the old families of the Eastern Seaboard. Children of immigrant 
parents are still sensitive concerning their full acceptance as first- 
class citizens, and tend to react by demonstrating a hyper-patriotic 
outlook. And a large group in American society has been able to 
rise in economic and social status as a result of industrial prosperity 
and the white-collar explosion, yet they often feel neglected when 
power and privileges are bestowed. This analysis is an imaginative 
one, and Parsons has a clear view of the sources and manifestations 
of serious strains in American life. He proceeds to show how these 
unrelieved tensions provided a large, if miscellaneous, constituency 
of support for McCarthyism. Compulsive concern about loyalty 
and security, treason and subversion, and about the softness of 
traditional leadership and the need for hardheaded measures— all 
of these were not passing political phases, but "symptoms of a proc- 
ess in American society of some deep and general significance." f 
To write this way is to write of a fundamental social disequilib- 
rium. McCarthy himself has passed from the scene. And McCarthy- 

* Talcott Parsons, "Social Strains in America," reprinted in Daniel 
Bell, ed., The New American Right (New York: Criterion Books, 
1955), pp. 117-18. 
f Ibid., p. 117. 


ism has either subsided or been institutionalized in our social 
structure and internalized in our personalities. The importance of 
Parsons' essay lies in its discussion of important social forces of 
which McCarthyism was only a symptom. The alleviation of symp- 
toms—in this case the censure of McCarthy— must never be con- 
fused with fundamental cure. It is of some interest that Parsons 
has not returned, since McCarthy's demise, to a consideration of 
the strains he so well outlined in 1955. If they are as deep-seated 
as he made them out to be, they cannot be ignored once their most 
disruptive symptoms have declined. It will therefore be worth the 
time to refer to some of the questions that Parsons raised. The 
McCarthyite constituency to which Parsons alluded consisted, on 
the whole, of two major groups : the successful and the unsuccessful, 
the upwardly and downwardly mobile. At the same time the 
movement's supporters may be divided into his vocal and virulent 
supporters, on the one hand, and those who gave him their tacit 
consent, on the other. The individuals who should be given careful 
attention are the successful Americans who offered their silent 
approval to the McCarthy crusade. There is no disputing that in 
the postwar years millions of individuals have experienced a rise 
in status. They have moved out of old neighborhods; they have 
put on white collars; they have been able to surround themselves 
with material comforts; and they have created a new image of 
themselves and new expectations for their children. What has also 
occurred is that these Americans have begun to take seriously their 
status as first-class citizens. This development is more startling 
than it might at first glance seem. For 150 years the American 
creed talked the rhetoric of equality, but these sentiments were 
never expected to be taken at face value by the great part of the 
population. Now, however, new millions of Americans are in a 
position to demand that equality. They are no longer illiterate 
immigrants huddled in the urban slums, they are no longer marginal 
farmers forgotten in the rural countryside. They are now American 
citizens: middle class and not a little arrogant about it. Problems 
arise because there has been a political lag in the course of this 
social advance. The promises inherent in the rhetoric of political 
equality have not been fulfilled, or not delivered to the extent 
that fibrst-class citizens have come to expect. In political terms the 

294 • Andrew Hacker 

emerging middle class remains relatively powerless. It is unorgan- 
ized, inarticulate, and incapable of promoting its political interests. 
Indeed, its "interests" are so generalized and inchoate that it is 
hard to know where to begin securing them.* 

These people, although Parsons did not say much about them, 
were the real supporters of the McCarthy movement. Too concerned 
with being respectable, they did not go to meetings, join organiza- 
tions, or write letters to the newspapers— either for or against him. 
It was their political silence and inactivity, however, which gave 
free rein to an era of demagogy. And it is now relevant to suggest 
that this constituency, fast approaching majority status in the 
country, will be the source of further strains. Only two will be 
mentioned here, but others may come to mind. 

The first is in the area of race relations, especially in the North. 
Things are going to get a lot worse, and it is not at all self-evident 
that they are going to get better in the foreseeable future. If there 
is one sword which hangs over the heads of untold millions of 
white— and northern— Americans it is that they cannot afford to live 
in close proximity to Negroes. The single social fact which can 
destroy the whole image of middle class respectablity is to be 
known to reside in a neighborhood which has Negroes nearby. 
Pollsters' notebooks are filled to overflow with the rationalizations 
supposedly impelling the flight: the danger of violence, overcrowded 
schools, not enough green grass and fresh air, and so forth.f But 
the simple answer is that these Americans are too insecure in their 
newly won status, too fearful of the opinions of others, too ready 
to take the easiest and available way out. Not simply our great 
cities, but all urban areas are developing racial ghettoes with 
inadequate social services and slender opportunities for escape for 
those who must stay behind. And our burgeoning suburbs have 
become monuments to white anxiety. The problem is a national 
one, and it is bound to become exacerbated as more white Ameri- 

* The tenn "interest" is being used here in the sense that James 
Madison intended in the Tenth and Fifty-First Federalist Papers. For 
a further explanation see Andrew Hacker, Politics and the Corporation 
(New York: Fund for the Republic, 1958), pp. 4-11. 
f See the Report of the Commission on Race and Housing entitled 
Where Shall We Live? (Berkeley and Los Angeles: University of 
California Press, 1958). 


cans are drawn into the middle class. For this class-status is too 
easily attained, too unstructured, to give those who enter it a 
sense of psychological security. The decision to move to a suburb is 
no solution to the basic problem. The answer must be political 
and is yet to emerge. 

The second area of strain has to do with the quality of culture. 
At one time in our history the constituency for knowledge and 
serious learning was a small one. The proportion of the population 
which went to college, which read important books and periodicals, 
and which generally partook of high culture was comparatively 
minute. With such a small and appreciative clientele, disciplined 
standards could be both set and met. All of this is being altered, 
and for good reason. The citizens of the new social democracy 
demand not only a high school education, but also college admission 
for themselves and their children. And the simple fact is that most 
of these people— Fortune magazine calls them "the new masses" * 
—are not equipped for serious or disciplined learning. When culture 
has a small, selective, and privileged constituency, it is possible to 
keep standards high: as the constituency is enlarged to many times 
its original size, the distribution of aptitudes and motivations is 
bound to be far wider and the mean far lower. Nevertheless, these 
new citizens demand admission to the citadels of knowledge, and 
once they are there they pull requirements down to a level they 
can handle. The point, in short, is that the new middle class is 
too large and too poorly motivated to live by the traditional injunc- 
tions of quality. Arguments about the number of "classical" records 
and local symphony orchestras, about the number of "good" paper- 
back books and local little theatres, are more wishful thinking than 
serious analysis. f American culture is increasingly yielding to 
majority wishes, increasingly being defined in mass terms. Even the 
most venerable of schools and universities cannot but be swayed by 
the demands of a buyers' market. This, then, is another consequence 
of making real the doctrine of equahty of opportunity. Social 

* Daniel Seligman, "The New Masses," Fortune, 59:106 S. ( 1959). 
f See Bernard Rosenberg and David Manning White, eds., Mass 
Culture (Glencoe, 111.: Free Press, 1957), especially the essays by 
Rosenberg, Dwight MacDonald, and Melvin Tumin, at pp. 3-12, 
59-73, and 548-56 respectively. 

296 • Andrew Hacker 

democracy and cultural majority rule produce strains less virulent 
than McCarthyism and less noisome than the race discrimination, 
but serious enough to warrant attention. And these tensions, too,, 
ultimately have a political content. 

It is no criticism of Parsons to point out that since the time of 
McCarthy he has not written articles on other strains in American 
life. But what is of interest is that since that time he has all but 
forgotten the structural factors which underpinned his analysis of 
the McCarthyite tensions. It has been suggested that these forces 
still exist and that they will continue to manifest themselves for 
a long period to come. And the key question for theory, of course, 
concerns what is going to emerge in the future. The writings of the 
great political theorists had two characteristics. On the one hand 
they were startling: they told us something new and unorthodox 
about the society we thought we understood. And on the other 
hand they stuck their necks out: they ventured a prediction about 
the future direction of social and political development. Parsons' 
work, if it is to have lasting value, must be assessed on both of 
these grounds. 

There is, first, the question of social class. In a paper entitled 
"Social Classes and Class Conflict," delivered before the American 
Economic Association, he offered a critical and yet sympathetic 
analysis of some features of Marxian theory. After examining the 
strengths and weaknesses of Marx's approach. Parsons then pro- 
ceeds to his own consideration of class conflict in modern, indus- 
tralized societies. The root of the matter, as he sees it, lies in the 
tension between the emphasis on individual attainment and the 
imperatives of bureaucratic organization. "The status of the indi- 
vidual," he says, "must be determined on grounds essentially 
peculiar to himself, notably his own personal qualities, technical 
competence, and his own decisions about his occupational career 
and with respect to which he is not identified with any solidary 
group." (Essays II 327) At the same time there arises a complex 
of organizational structures which have the power to direct signifi- 
cant elements in the lives of individuals. "Organization on an ever 


increasing scale . . . naturally involves centralization and difiFer- 
entiation of leadership and authority," Parsons says, "so that those 
who take responsibility for coordinating the actions of many others 
must have a diflFerent status in important respects from those who 
are essentially in the role of carrying out specifications laid down 
by others." (Essays II 327) Apart from the empirical question of 
how many individuals are affected by these organizational impera- 
tives and to what degree, there is little to argue about in these 
descriptive propositions. There then follows a listing of the "prin- 
cipal aspects of the tendency to develop class conflict in our type 
of social system." (Essays II 329-32) These may be summarized: 
(1) In a competitive occupational system there will be losers as 
well as winners. (2) Organization entails discipline and authority, 
and there will be resistance to the exercise of this power. ( 3 ) Indi- 
viduals favored by strategic location can exploit those less fortu- 
nately placed. (4) Varied and conflicting ideologies emerge in a 
differentiated social structure. (5) Patterns of family life and 
attitude-formation in the young will vary as between social classes. 
( 6 ) The promise of equal opportunity for all will be thwarted. 

Parsons acknowledges his indebtedness to Marx wherever appro- 
priate (and it would be pleasant if more social scientists were 
secure enough to be able to do the same), and he also quite 
properly indicates that Marx's theory is insufficient to explain the 
contemporary world. Indeed, what social science most surely needs 
is a new Marxism: a new systematic theory which postulates cause 
and effect and which commits itself on the future development of 
society. Such a theory, however, needs what for lack of a better 
term may be called a source of energy. The common criticism of 
Marx is that he had but a single, determinist idea at the core of 
his thinking. But at least it was an idea of some power, and he 
was able to develop the rest of his thoughts around it. The diflS- 
culty with Parsons' scheme is that he has too many ideas which 
interact on a parity of causal significance. It might be asked which 
one of the six "principal aspects" of class conflict is most important, 
which one— if any— is causal with relation to the rest. One is 
tempted to conclude that until Parsons is prepared to be a little 
less conventional, a little more daring, we will not have a pioneer- 
ing explanation of social strains or class conflict. We might, indeed. 

298 • Andrew Hacker 

ask whether the social strains America has been experiencing are 
instances of class conflict in modern dress. The new middle class 
has many of the attributes of an alienated proletariat, albeit a 
proletariat with white collars. However, there is lacking a class- 
consciousness in any political sense; and the exploitation of a 
bourgeois class has been replaced by the discipline and authority 
of impersonal corporate institutions. Many important political ques- 
tions are raised here, and it may be hoped that Parsons will turn 
to them before long. 

One obstacle to a Parsonian theory of class and power may not 
be easy to overcome. Social scientists, whether they acknowledge it 
or not, cannot help being bearers of an ideology— although the 
ideology will, of course, diflFer from person to person. Ideology, 
for present purposes, may be thought of as having two components. 
It is, first of all, purportedly normative, composed of philosophical 
propositions which are actually rationalizations for preserving the 
status quo or attaining a new set of social arrangements. Second, 
ideology is purportedly scientific: an unintentionally distorted pic- 
ture of social reality, the distortion arising because the observer 
sees what he wants to see. Any theory which combines fact and 
norm, whether by accident or design, runs the risk of forcing 
descriptive reality into the Procrustean bed of ideology. This is 
probably inevitable, and it is certainly not to be condemned out 
of hand. Indeed, the real test is not whether fact or norm is 
tainted with ideology, but whether the ideology itself is a viable 


The ideological overtones in Parsons' political thinking come to 
light most vividly in his essay on " 'Voting' and the Equilibrium of 
the American Party System." This is ostensibly a review-article, 

* Jeremy Bentham put it this way: "No wonder then, in a treatise 
partly of the expository class, and partly of the censorial, that if the 
latter department is filled with imbecility, symptoms of kindred weak- 
ness should characterize the former." A Fragment on Government, 
edited by Wilfrid Harrison (Oxford: Blackwell, 1948), p. 14. 


drawing on the Elmira study of Berelson, Lazarsfeld, and McPhee.* 
Actually, however, Parsons uses only those data which are helpful 
to his own analysis of the party system in the United States; and 
he gives no evidence of a familiarity with the voluminous literature 
on party politics which has accumulated in the postwar years. His 
particular interest in the process of representative government as 
it relates to social stability, "My point of reference will be the 
capacity of a social system to get things done in its collective 
interest," Parsons says. "Hence power involves a special problem 
of the integration of the system, including the binding of its units, 
individual and collective, to the necessary commitments." f It is 
Parsons' view that things do get done through the medium of the 
party system and that the system does remain integrated. To con- 
clude this, however, is to make a number of important assumptions: 
about what ought to be done, what can be done, and the effective- 
ness of what is done. There is, furthermore, the assumption that 
what we see at work is actually the process of representative 
government. To begin with the last of these, Parsons believes that 
the institutions of political democracy play an important and effec- 
tive role in the exercise of power in society. The chief of these 
institutions is the vote as it is exercised through the party system. 
"Voting is the central focus of the process of selection of leadership 
and hence in one sense all other influences must channel their 
effects through the voting process," he says. "The two-party system 
may be regarded as a mechanism that makes possible a certain 
balance between effectiveness through a relative centralization of 
power, and mobilization of support from different sources in such 
a way that . . . the supporter is offered a real alternative." J While 
this description of voting and the party system in America is admi- 
rable from the standpoint of normative democratic theory— the 
writings of Robert Maclver or Ernest Barker, for example— it bears 
small relation to how these institutions operate. 

* Bernard R. Berelson, Paul F. Lazarsfeld, and William N. McPhee, 
Voting: A Study of Opinion Formation in a Presidential Campaign 
(Chicago: University of Chicago Press, 1954). 

f Talcott Parsons, " 'Voting' and the Equilibrium of the American 
Political System," Eugene Burdick and Arthur J. Brodbeck, eds., 
American Voting Behavior (Glencoe, III.: Free Press, 1959), p. 81. 
I Ibid., pp. 86, 87. 

300 • Andrew Hacker 

The point is not that Parsons has got his facts wrong: actually 
the interpretation of reality is far from settled in this area. What is 
important is that Parsons has come to his particular interpretation 
and that he has seen fit to reject other alternatives. And it follows 
that he has chosen to emphasize certain facts and to ignore others. 
The question, to repeat, is why he sees what he does and why 
he turns a blind eye in other directions. A few comments on matters 
of fact— or the interpretation of fact— may be in order. The selection 
of leadership by means of the vote, it may be argued, only assumes 
significance in limited cases: if the two individuals on the slate 
of candidates selected by the two party organizations offer a real 
choice in terms of their stands on matters of policy. Usually they 
do not. In most parts of the country the bearer of the party label 
traditional to that area wins automatically. In most contested dis- 
tricts the tendency is for candidates to be essentially similar because 
both must appeal to the same heterogeneous electorate. On the 
national scene it is possible to claim that American voters have 
not been offered a "real alternative" since Bryan ran against 
McKinley in 1896. (It turned out that they had a real choice in 
1932, but the voters did not know it while the campaign was 
going on.) Furthermore, it is quite plausible to suggest that the 
major interests which exercise an influence in the making of public 
policy make their weight felt regardless of the candidates the 
voters happen to put in office. Such interests are studiously non- 
partisan and they are quite ready to approach policy-makers no 
matter which party label they happen to bear. These are only a 
few alternative interpretations of the voting and party processes, 
and space forbids elaborating on these or others at this time. 

The reason why Parsons presents such a one-sided picture can 
only be a subject for speculation. His chief concern, it would ap- 
pear, is to show that the American political system is a democratic 
one at base. He wishes to present a persuasive case to the effect 
that the public has power and that it uses this power to govern 
itself. The voter, in short, can use his ballot as an instrument for 
compelling his rulers to make policy responsive to his wishes. "He 
receives the expectation that many kinds of measures that he 
approves will be implemented if his candidate wins, but without 


exact specification of particular policies," Parsons says.* But even 
this carefully qualified statement, it may once again be suggested, 
describes the ideal rather than the real. Voters continue to expect 
that promises will be delivered— their faith, although occasionally 
tinged by cynicism, is self-renewing— but even those who support 
the victors are usually disappointed. Taxes are not cut, the cost of 
living continues to rise, unemployment is never fully abolished, 
peace with honor remains an unfulfilled hope. And when it comes 
to even more subtle issues, issues between the lines of the formal 
platforms and speeches, our political institutions have shown them- 
selves incapable of rising to the challenge. Parsons, however, is 
content with what he sees. In terms of the mechanisms of repre- 
sentative government and in terms of the substance of public 
policy, he sets his standard for optimum performance at a fairly 
low level. "The essential point is that new things do get done and 
that the consequences do come to be accepted," he says. "In 
view of what sociologists now know of the intensity of the tensions 
and stresses generated by major processes of social change, the 
relative effectiveness of this set of mechanisms is impressive." f 
What makes them look impressive is that Parsons believes they 
have been subjected to a rigorous test and have passed that exam- 
ination successfully. 

An example of this testing is the New Deal, with business regu- 
lation and social welfare legislation. "The Federal Reserve Act, 
the Securities Exchange Act, the Wagner Act, and the Social 
Security Act, were all Democratic measures— every one of which 
was strongly contested by the Republican party," Parsons says. 
"Every one of them has come to be fundamentally accepted by 
that party with no attempt to undo the work." J As a factual 
proposition this is of course true. What Parsons finds impressive 

* Parsons, " 'Voting' and the Equilibrium of the American Party 
System," p. 90. 
f Ibid., p. 112. 

I Ibid., p. 111. The "Democratic party" which Parsons refers to in 
generahzed terms is actually the liberal— and minority— wing of the 
party. Like most Uberal Democrats, Parsons would like to believe 
that his image of the party is the real one. The record of the Demo- 
crats in the Congress over the past twenty years, however, shows that 
the reality lies elsewhere. 

302 • Andrew Hacker 

is the fact that the Repubhcans accepted these laws, that the 
business community in particular did not resort to extra-constitu- 
tional means when they were put on the statute books. An alterna- 
tive view would suggest that the limits of the American political 
consensus has not been tested since the close of the Civil War. 
What Parsons and other liberals like to think of as business regula- 
tion is, despite the predictable complaints of businessmen, more 
a paper tiger than an effective system of economic controls in the 
public interest. A few questions may be asked about these sup- 
posed powers of the national government. Can any public agency 
determine the level of wages, of prices, of profits? Can it, perhaps 
more important, specify the level and direction of capital invest- 
ment? Can any government bureau allocate raw materials or con- 
trol plant location? Can it in any way guarantee full employment 
or the rate of economic growth? Has any suit of the Anti-Trust 
Division actually broken up one of our large corporations in an 
appreciable way? The simple answer is that measures such as 
these are neither possible under the laws nor do we know what 
the reaction to them would be. And what Parsons chooses to call 
welfare legislation is, despite the partisan panegyrics of New Deal 
Democrats, more a humane hope than a realized system of eco- 
nomic security. Several questions may once again be posed. What 
proportion of low-income Americans live in rural or urban slums 
and what proportion are in government housing projects? What 
source of income is there for a man who is out of work after his 
13 or 26 weeks of unemployment compensation expires? What 
standard of life can be maintained on the social security pension 
an individual may receive at 65 and how many Americans do not 
have additional sources of income? If a family is visited with a 
really serious and extended illness, where can a citizen get medical 
care other than in a charity ward? Just what can a widow or a 
deserted mother with three small children expect as a right from 
her government? Any serious stLidy of these matters will show 
that the so-called welfare state offers a slender mite ideed. 

In making judgments in an area like this, one can be a Burke 
or a Bentham, but in neither case is one a social scientist. To a 
Burkean what has been done looks impressive; to a Benthamite 
what remains to be done looks formidable. Parsons is pleased with 



what has been accomphshed: to his mind it is quite a feat that 
so much has been done without rending the Repubhc. His evidence 
that the hmit has been reached is that businessmen complained so 
bitterly about even minimal regulation and welfare measures. This, 
it may be suggested, is no test at all. Businessmen complain without 
surcease, and have been doing so since the time of Adam Smith.* 
We do not know how much they will take without resorting to 
counterrevolution. Parsons' political "equilibrium" is, on the one 
hand, an acceptance of the economic status quo in its major out- 
lines, and, on the other, a cautious espousal of traditional liberalism. 
The latter will be examined more carefully later on. It might also 
be asked whether the government is in a position to do anything 
about the "tensions and stresses generated by major processes of 
social change" which Parsons himself has discussed. Here the 
focus is not on regulatory or welfare problems, but on the larger 
social strains. If it is not government's, then whose responsibility 
is it to remedy the status anxieties, the fragmentation of personality, 
and the sense of individual powerlessness brought on by contem- 
porary institutions and events? 

Parsons has said that the instrumentality of the vote is important, 
but surely there are limits to solving such problems via the ballot 
box. The forces which led to McCarthyism were not exorcised by 
the censure and death of McCarthy. Racial tensions will not be 
solved by pleas for tolerance, and the cultural level will not be 

* More than a century ago Charles Dickens could report: "Surely 
there never was such a fragile china-ware as that of which the millers 
of Coketown were made. Handle them never so lightly, and they feU 
to pieces with such ease that you might suspect them of having been 
flawed before. They were ruined, when they were required to send 
labouring children to school; they were ruined when inspectors were 
appointed to look into their works; they were ruined, when such 
inspectors considered it doubtful whether they were quite justified 
in chopping people up with their machinery; they were utterly un- 
done, when it was hinted that perhaps they need not always make 
quite so much smoke. . . . Whenever a Coketowner felt he was ill- 
used— that is to say, whenever he was not left entirely alone, and it 
was proposed to hold him accountable for the consequences of any 
of his acts— he was sure to come out with the awful menace, that he 
would 'sooner pitch his property into the Atlantic' This had terrified 
the Home Secretary within an inch of his life, on several occasions." 
From Hard Times (1854). 

304 • Andrew Hacker 

raised by pleas for internal discipline. Our political institutions, 
as now constituted, are too free and too democratic to handle these 
problems of status and personality. Serious questions must be 
raised about democracy and freedom. What is revealing is that 
Parsons evades these questions altogether. The solutions provided by 
communism and fascism are abhorrent; those proposed by classical 
conservatism and socialism are pre-industrial and hence Utopian. 
There is nothing wrong with talk of "equilibrium" if its base point 
is reasonably up to date. Parsons may, like many of us, be fearful 
of what the future will bring. But that is no excuse for designing a 
political theory which stands still. 

Parsons' nostalgia for the past and his acceptance of present 
arangements are brought out most clearly in his article-review of C. 
Wright Mills' The Power Elite."* Mills' book, one of the most chal- 
lenging to appear since the end of World War II, speaks a language 
which is harsh and alien to Parsons' ears. It is interesting to see 
what Parsons makes of these arguments, for in a real sense we have 
here a confrontation of liberal and radical thinking. It is not 
surprising that Parsons fails to understand much of what Mills has 
to say. The discussion of who the members of the power elite are 
is neglected in order to make the rather obvious point that America 
is no longer ruled by a property-owning class. And as for Mills' 
important chapter on mass society, Parsons thinks it has to do 
mainly with mass media and he does not know what to make of it. 
There are comments on the role of women and physicians (they 
are socially important ) ; on government economic controls ( they are 
genuine because businessmen object to them); on Adlai Stevenson 
(a favorite of Parsons'); and on the fact that Americans have 
friends and relations and go to church ( so they cannot be as anomic 
as is made out ) . Finally Parsons says that he is not really interested 
in how power is distributed— who exercises it over whom and who 

* C. Wright Mills, The Power Elite (New York: Oxford University 
Press, 1956). 


has it at whose expense— but rather in how it is produced. "Power 
is a generahzed facihty or resource in the society," he says. "It has 
to be divided or allocated, but it also has to be produced and it 
has collective as well as distributive functions." * It is clear that 
Parsons, for the symmetry of his larger social system, wants to set 
up a wealth-power analogy in order to underpin an economics- 
politics model. The scheme dictates that if wealth must be created 
before it is distributed, so must power. Until Parsons can show that 
it is important to make this analytic separation, it will probably be 
better to follow men like Machiavelli and Hobbes who find the 
production and distribution of power an identical process. For Par- 
sons, however, the dichotomy serves the useful purpose of allowing 
him to evade the controversial questions raised by asking who in 
America has power and who has not. 

Mills' book is more complex than it seems on the surface and is 
not easy for someone reared on liberalism to understand. Parsons 
understands that something akin to a power elite exists. "The rise 
to prominence within the firm of specialized executive functions," 
he says, "is a normal outcome of a process of growth in size and in 
structural differentiation." f This is true enough. But Mills' concern 
is with the great influence that the decisions of these top executives 
have on the lives of Americans, a power in no way made institu- 
tionally responsible. Parsons skirts this problem, and in doing so 
implies that he does not think it important. And the idea of the 
mass society, the other side of Mills' theory, receives even less at- 
tention from Parsons. There are millions of Americans— the Ameri- 
cans described in White Collar, The Organization Man, The Lonely 
Crowd, and The Status Seekers— who have no significant access to 
power.J To Parsons' mind they have the vote, and this makes them 
masters of their destiny. Mills juxtaposes the anonymous and non- 
responsible men who lead the great corporate institutions and the 

* Talcott Parsons, "The Distribution of Power in American Society," 
World Politics, 10:141 (October 1957). 
f Ibid., p. 129. 

I "No social scientist has yet come up with a theory of mass society 
that is entirely satisfying," Irving Howe says; but he himself gives a 
cogent description of its bare outlines. See his "Mass Society and 
Post-War Modem Fiction," Partisan Review, 26:426-28 (Summer 

306 • Andrew Hacker 

cheerful and anxious Americans who are recipients of commands. 
All this and more is in Mills' book, but Parsons is unable or will- 
ing to see it. Mills is not so much describing the present as he is 
picking out future tendencies. Because Parsons has no view of the 
future himself, he can only quarrel over details. And he can also 
criticize Mills' tone. "There is," Parsons says, "the tendency to think 
of power as presumptively illegitimate; if people exercise con- 
siderable power, it must be because they have somehow unsurped 
it where they had no right and they intend to use it to the detriment 
of others." * When Mills finds both irresponsibility and immorality 
in the conduct of the power elite. Parsons becomes a realist and 
decries JeflFersonian Utopianism. What is required, he says, is "ob- 
jective analysis." 

We have now come full circle. C. Wright Mills is called a 
Utopian because he would prefer it if a power elite did not exist. 
The question which now has to be put is why Parsons prefers that 
McCarthyism and the power behind it not exist. "McCarthyism," he 
says, "is perhaps the major type of 'pathology' of our system and, 
if not controlled, may have highly disruptive consequences." f If 
McCarthyism is "pathological," why not also the Higher Immorality 
of a power elite? The use of a clinical term can be deceptive. (Are 
race prejudice and mass culture also "pathologies"? What about 
labor disputes, juvenile delinquency, isolationism in foreign aflFairs, 
and the dearth of good conversation?) It is clear that Parsons is 
assuming that certain social arrangements are healthy and others 
are not. A good idea of his conception of normality may be gained 
if we look at his prescription for the McCarthyite disease. Power 
must be met with power: in this case the power of the populace 
with the power of— yes— the power elite. "Under American condi- 
tions, a politically leading stratum must be made up of a com- 

* Parsons, "The Distribution of Power in American Society," p. 140. 
f Parsons, " 'Voting' and the Equilibrium of the American Political 
System," p. 103. 



bination of business and non-business elements," Parsons says. "The 
role of the economy in American society and of the business element 
in it is such that political leadership without prominent business 
participation is doomed to inefiFectiveness and to the perpetuation 
of dangerous internal conflict. It is not possible to lead the Ameri- 
can people against the leaders of the business world." * Parsons 
suggests that business leaders be brought into politics and that they 
use their social power to quash the popular attack on civil liberties. 
This prescription probably reveals better than anything which has 
been said up to now Parsons' view of political and social normality. 
It is his hope that the men in the higher ranks of America's cor- 
porate world are potential defenders of the traditional liberal 

Historically speaking, such a view of the business class is justified. 
The growth of political liberty in the Western world was accom- 
panied by, even caused by, the ascent to power of men of property. 
This class was informed that wisdom, virtue, and social responsi- 
bility were its proper attributes; and in many ways it lived up to 
these expectations. Its members put their power and prestige be- 
hind the Constitution, the Bill of Rights, and the Common Law. 
They were able to do this because they were accorded unques- 
tioning deference by a public which acknowledged that their betters 
ought to attend to matters as important as these. The source of 
this class' power lay in its property ownership and its members' 
personal control over elements of the economy. They also supplied 
the nation's diplomats and cabinet members, lawmakers and judges, 
financiers and industrialists, churchmen and scholars. While this 
class was the custodian of the country's liberty, it was careful not 
to overextend its resources in their defense of the rights of in- 
dividuals. The railroaded Wobbly in Montana and the emasculated 
Negro in Alabama were not encompassed by their power: and they 
could not find it in their jurisdiction to defend two Italian anarchists 
named Sacco and Vanzetti. By and large the freedoms which this 
class created were for their own use; but they were phrased in 
universalistic and equalitarian terms, and there was a residue for 
the rest of society. This class, also, was the major support of higher 

* Parsons, "Social Strains in America," p. 139. 

308 • Andrew Hacker 

learning and serious culture: here too it was for their own benefit, 
but the standards stood for the community as a whole. The need 
for such a ruling class is implicit in Parsons' notion of political 
equilibrium, although it is doubtful if he would acknowledge it— or 
if he even realizes it. What is "pathological" about McCarthyism— 
and racial discrimination and mass culture— is that the man in the 
street is no longer deferring to his betters. In his essay on voting 
Parsons says, "In constructing this model I have of course leaned 
heavily on the literature of political theory." * That literature, from 
Plato and Aristotle through Locke and Mill, relies on the power of 
a secure ruling class to protect the traditional liberties of a society. 
Scholars who have benefited from this shield, who reside in in- 
stitutions which continue to feed on prescriptive deference, may be 
excused if they generalize from their particular good fortune. f 

This America, the creation of the liberal ideology and class struc- 
ture, is passing rapidly from the scene. Already the old class has 
had to share its power with the new elite. This is an elite of talent, 
but specialized talent. They are the men Mills excoriates and the 
men Parsons calls on to take up the defense of political freedom. 
However, if any examination is given to the kind of men they are, 
their interests in life and their social backgrounds, the basis on 
which they were selected and their own definition of their roles, 
and above all to their unwillingness to entangle with controversy— 
if such study is made, it soon becomes apparent that these men 
have neither the concern nor the motivation to use the power of 
their institutions to defend the freedoms so cherished by traditional 
liberalism. They define their responsibilities to society in only the 
most cautious and conventional of terms. For all the rhetoric about 
"the conscience of the corporation" and "the social responsibilities 
of business," when the chips are down the elite has shown itself un- 
willing to oppose the pathological strains which Parsons deplores. 

* Parsons, " 'Voting' and the Equilibrium of the American Political 
System," p. 113. 

f For a further development of the ideas in this paragraph and the 
one following see Andrew Hacker, "Liberal Democracy and Social 
Control," The American Political Science Review, 51:1009-26 (De- 
cember, 1957). 


While this might be expected of the old ruling class, it is quite an- 
other thing to ask such deeds from the new elite. They simply 
would not understand what Parsons is talking about. 

Concurrently with the rise of the elite, the children and grand- 
children of a once deferential public are beginning for the first time 
to feel their democratic oats. The democracy is more social than 
political, but its consequences cannot be ignored. Experiencing ten- 
sions as they move into a place in the sun, these people are com- 
pelled for the benefit of their own well-being to act in ways that 
are inimical to traditional freedoms. Overt political populism is only 
sporadic: the defense of a white neighborhood in one instance, the 
defeat of a school bond issue in another, obscurantist legislation 
from time to time as a third. But society is too bureaucratized for 
populist politics to damage the structure itself: what is far more 
fragile is the delicate fabric of traditional liberty, and here the cost 
can be high. The new masses, furthermore, have no vested interest 
in such protections as the First Amendment freedoms. For them 
freedom is not the right to make a heretical speech, but the right 
to move to the suburbs and buy a motorboat. The new and 
burgeoning middle class, unlike the old and selective middle class, 
is without commitment to political liberty or a culture of quality. 
And the new elite, while it exercises control over much of the 
economy and society, makes no eflFort to contain the "pathological" 
behavior of the new democracy. The chief explanation for this is 
that elite and mass are really not much different so far as tastes or 
interests are concerned: the former simply have more important 
jobs than the latter. Politically and culturally they are quite similar. 
Both subscribe to Life magazine. 

The ideology underlying Parsons' political theory is a worthy one 
in many respects. But liberalism of the eighteenth and nineteenth 
centuries no longer has the structural basis which gave it its 
strength. The political era into which we are moving will create 
its own equilibrium: and both a power elite and a mass society will 
play crucial roles in its definition. Neither Mills' socialism nor Marx' 
communism, and least of all Parsons' liberalism, will be of much 
theoretical help. The ideological components of Parsons' thought 
do him a disservice not because they are ideological, but because 

310 • Andrew Hacker 

they refer to a world we have put behind us and to which we can- 
not return. The poHtics he depicts are the pohtics he would like to 
exist, not those we are going to have to Hve with. Whether this 
nostalgia for an age of civility infuses the larger outlines of his 
social system is a question that all students of Parsons ought to 



Talcott Parsons 








It is a matter o£ great satisfaction to an author to have 
the kind o£ serious and competent attention to be paid to 
his work of which the present volume is the expression and 
product. It is not only the personal honor (which, how- 
ever, I greatly appreciate) which is the source of satis- 
faction, but also the advancement of the cause to which 
we are all committed, of furthering the development of 
scientific theory in the field of human behavior. The very 
existence of such a volume is to my mind an important 

312 • Talcott Parsons 

index of the development of an increasingly mature state of the 
relevant disciplines. 

The nine essays which have preceded this one, with their gener- 
ally high level of critical competence and seriousness, are also inter- 
esting for the fact that they do not present any unified critical point 
of view. They do, of course have a unity, consisting in common 
orientation to the body of work under consideration, but their 
authors direct attention to so many diflFerent aspects and are both 
attracted and bothered by so many different features of it that sub- 
stantively one can perhaps call the main common focus a deep 
concern for the problems of social science, a concern which uses 
this body of material as a reference point. 

Confrontation of an author with these essays naturally stimulates 
reflection both on relevant aspects of his personal intellectual his- 
tory and of the place of the subject matter in the cultural situation 
of the time. Perhaps I may start with a few observations on the 
latter topic, then proceed to a few on the former, and from these 
pass on to a few essential problems for their own sake. 

One of the indications of the intellectual situation of our time is 
the prevalence of self-consciousness about what is going on; one 
example of tliis in turn which closely touches our own field is the 
development of the concern and to some extent the discipline 
known as the sociology of knowledge. This has made us acutely 
aware of the immense variety of patterns according to which con- 
ceptual schemes which have eventually turned out to be scientifi- 
cally important, have been received, both in professional circles and 
among the more general public. 

It is almost a commonplace that ideas which entail a major re- 
organization of patterns of thinking in their field are very likely 
to encounter severe opposition. This is perhaps the more true when 
they are ideas directly involved in the structure of the society and 
culture in and by which their authors live. Indeed, it seems to me 
that this is very relevant to the facts, first that it is in the socio- 


cultural field that genuine science has tended to develop latest, 
and second that it is in such a very controversial phase at the 
present time. 

Certain major scientific innovations, however, have from the be- 
ginning been acclaimed, at least within the relevant professional 
groups— important cases would be the Newtonian and Einsteinian 
theories. Others, like Mendel's work in genetics, have for long 
simply failed to excite any interest at all. Still others, like Pasteur's 
theory of the role of infecting agents in disease, have stirred up 
violent controversy and sharp repudiation by at least a large part 
of the lelevant professional groups. 

In such cases it is understandable that an important part of the 
opposition stirred up— as well as of the support— should have ideo- 
logical as well as scientific components. This again is particularly 
true of social science since it touches so closely the value commit- 
ments and other cultural vested interests of contemporary groups. 

Perhaps some interpreters of the situation would differ, but it 
seems to me to be broadly correct to state three points about the 
situation of social science. The first is that the most massive founda- 
tions for the development of modern social science were laid in 
England in what can be broadly called the utilitarian movement of 
the 18th and 19th centuries— with important Continental connections, 
especially with France. To this impulse may be attributed not only 
the firm establishment of economics, and the more "Benthamite" 
tradition of political science, but also the foundation of modern 
anthropology by Tylor and the very important sociological perspec- 
tives of Spencer. 

Second, however, the major breakthrough into the perspec- 
tives which supported the development of the "behavioral sciences' 
in the last generation or two came, in terms of theory, not from Eng- 
land, but from Continental Europe, the major figures being Durk- 
heim and Max Weber on the one hand, Freud on the other. This 
involved contact with the "idealistic" and collectivistic components 
of the traditions of Western thought which were on the whole un- 
congenial to the British cast of mind. There was an important, 
though secondary, breakthrough in the United States with the 
"social psychology" of such figures as Cooley, Mead and W. I. 
Thomas. Also very important is the extension here of a basically 

314 • Talcott Parsons 

utilitarian pattern of thought in the great advance of experimental 
psychology in the early part of this century. 

In a particularly interesting way the United States has turned 
out to be the principal location in which the development from 
these major points of orientation has come to focus. It has above 
all seen an establishment of the behavioral sciences on profes- 
sional levels which have met with appreciable counterparts in 
Europe and those largely under American influence only very re- 
cently. The most important phenomena here have been the "draw- 
ing up" of sociology, anthropology, and psychology as disciplines to 
a place fully equal to those of economics and political science in 
the academic hierarchies; their full establishment in the structure 
of the universities; and the increased number of qualified and pro- 
fessionally trained personnel.^ 

Intellectually, this important American development has been 
"typically" American in that it has been relatively "pragmatic" to 
the point of often appearing to be eclectic. "Schools," like behavior- 
ism in psychology and the Sumner-KeUer cult in sociology have 
tended to be short-lived. There has been a strong emphasis on 
empiricism which has motivated a high concern, in all related fields, 
with the development of new research techniques. 

With respect then to anything like general theory, the situation 
has become complex. The American intellectual scene has been 
characterized by a marked openness and receptivity. Here it may 
be contrasted with what was, until rather recently at least, marked 
British traditionalism, a tendency to hold that none of the "new- 
fangled" theory could possibly be important, and at the same 
time a tendency widespread on the Continent for the problems to 
be defined predominantly in ideological terms, so that genuinely 
technical theoretical discussion has been blocked. Certainly, how- 
ever, many particular tenets of the new theoretical corpus have 
been received and developed in research, as, for example, Durk- 
heim's concepts of organic solidarity and anomie, Weber's concern- 
ing bureaucracy, and much in the work of Freud, including its 
relevance to the borderlines of sociology and anthropology. . 

^ I have tried to delineate this picture in some detail for sociology in 
the paper, "Some Problems Confronting Sociology as a Profession," 
American Sociological Review, August, 1959. 


Two further very important aspects of American openness have 
been a relative immunity to the pressure to put problems in an 
ideological context, i.e., a readiness to deal with problems of social 
science, and the related willingness to consider relatively particular 
and restricted contributions on their merits without worrying too 
much about the more "global" problems which lay in the back- 
ground. One could thus, for example, under this approach consider 
Durkheim on anomie in relation to suicide without worrying too 
much, in the Continental intellectuals' manner, about the special 
implications of his version of positivistic philosophy. 

The other side of this favorable picture has of course been the 
general American skepticism about high levels of generalization. 
Hence it has, in a certain way, been an unfavorable intellectual 
climate for the development and propagation of highly general con- 
ceptual schemes, since the question is always insistently raised 
whether this is necessary or even desirable in any sense. There is 
then necessarily in the American situation a set of resistances to the 
attempt to work at the level of general theory at all. In the present 
set of essays it is brought out most vividly by Professor Morse in 
his very clear statement that the relatively established general 
scheme which has dominated recent economic theory, in this coun- 
try as elsewhere, has erected certain clear barriers against raising 
borderline theoretical questions. Comparable positions may be 
found in "orthodox" psychoanalytic theory and in some phases of 
anthropological "culture" theory. However, there is a certain rela- 
tivity about this. 

It is in terms of this very broad diagnosis of the intellectual situa- 
tion of social science in the United States, that I would like to say 
a few words about personal orientation.^ There is of course no 

2 There is a sense in which the following remarks should be inter- 
preted more as a kind of "retrospective teleology" than a purely his- 
torical account. It is a framework in which, after several decades of 
activity, I tend now to interpret certain aspects of the "meaning" of 
my work, rather than a circumstantial account of "how" it came 
about. With respect to an interpretation of the genesis and "strategy" 
of the theory of action, the reader may wish to compare "An Approach 
to Psychological Theory in Terms of the Theory of Action" in Sigmund 
Koch, (ed.) Psychology: a Science, Vol. Ill, New York: McGraw- 
HiU, 1959. 

316 • Talcott Parsons 

doubt that the main "goal" has been to contribute to the develop- 
ment and propagation of "general theory" in the field which I gradu- 
ally came to think of as the sciences of "action." Whatever factors 
of temperament or of education may have underlain commitment 
to such a goal, the primary immediate idea came from my ex- 
posure to the work first of Max Weber, then of Durkheim and 
finally of Pareto. This, furthermore, occurred in the context of a 
major interpretive problem, that of the relation between the main 
traditions of economic theory and the interpretation of many salient 
characteristics of modern industrial society. 

This basic interest crystallized in my doctoral dissertation at 
Heidelberg on the Concept of Capitalism, with special reference to 
the work of Sombart and Max Weber. A relatively clear distinction 
between the scientific and the ideological aspects of the problem 
was worked out fairly early, and primary attention given to the 
former. In this context it became very clear that the problem of 
empirical interpretation or "diagnosis" could not be adequately 
handled without attempting to make far more explicit than was 
ordinarily done the extra-economic theoretical framework within 
which economic theory would have to be made to fit. If properly 
approached this could be seen to be a major theme in the work not 
only of Weber, but also of Durkheim and, very explicitly, of Pareto. 
Having worked out this theme to a degree in the writings of these 
authors, with Marx in the background, I attempted to tackle it in 
the work of the most influential economic theorist of the generation 
spanning the turn of the century, Alfred Marshall. The putting to- 
gether of all these things eventuated in the book the Structure of 
Social Action (first pubHshed, 1937), which is the basic reference 
point of all my subsequent theoretical work ( it is only very casually 
mentioned in any of the above essays except that of Devereux). 

I bring up this first major work here because it is such an im- 
portant reference point, not only in terms of content, but also for 
what I may call the strategy of theory-building. The convergence 
which I was able to demonstrate in that study, between the broad 
conceptual schemes used by these four authors, constituted the first 
level of integrated general theory in my own work. This was clearly 
very far from being a logico-deductive theoretical system in the sense 


referred to by Professor Black, but equally clearly it was very much 
more than an eclectic collection of unrelated theoretical ideas. 

Economic theory, broadly at the level achieved by Marshall, was 
undoubtedly the most highly sophisticated theoretical scheme yet 
developed for the analysis of any phase of human behavior, cer- 
tainly on the macroscopic levels. To my mind the most important 
theoretical contribution of the Structure of Social Action was the 
demonstration of a systematic range of problems on the borderline 
of this theory, and a convergent body of concepts oriented toward 
dealing with these problems. The most fundamental of these was, I 
think, the problem of order. It included also the problem of "ration- 
ality" and the clarification of the two basic meanings of the con- 
cept, the "psychological" meaning of motivational components ac- 
counting for deviations from rational norms, and, on the other 
hand, the "cultural" concepts of values, "ultimate ends," and so forth, 
which were nonrational rather than irrational. These were all related 
to and underlay the conception of a normative order of institutions 
such as contract, property, authority, and so forth, and some recon- 
struction of the relation between these institutions and the "self- 
interest" which was the focus of the motivational conceptions in 
economic tradition. They included the anchorage of the "moral 
authority" of normative patterns in religious commitments as ana- 
lyzed in Durkheim's and Weber's concepts, respectively, of the 
sacred and of charisma. 

It might have been conceivable to work the level of convergence 
just discussed directly into a logically tight theoretical system. I am 
inclined to doubt that under any circumstances this could have 
been fruitful; it would have been even more premature than Pro- 
fessor Black thinks my later attempts at systematization to have 
been. In any case, it was not the path which was actually taken. On 
the contrary mine was an American type of pragmatic path. It was 
to take up a whole series of restricted problems dealing with aspects 
of the more general scheme, and to work on them with the double 
reference to their logical and theoretical structure and the available 
empirical evidence. 

The first of these was the problem of some implications of the 
economic doctrine of self-interest. The setting chosen was the struc- 
tural contrast between business and the professions in modem 

318 • Talcott Parsons 

society, a problem greatly neglected in the economically oriented 
literature about "capitalism" whether in ideological terms it was 
conservatively or radically oriented. This in turn led into questions 
which I would now phrase as those of the relations between per- 
sonality and role, especially involving what above was called the 
psychological aspect of the historic problem of rationality. This led, 
by way of study of medical practice, into the whole problem of the 
social control and eventually of the genesis of nonrational motiva- 
tion. It was in this connection that an intensive study of Freud was 
first undertaken and personality theory became a serious concern 
for this particular sociologist. It strongly reinforced my conviction 
that sociology, as one aspect of the theory of social systems, could 
not sucessfully deal with many of its borderline problems without 
theoretically systematic consideration of many of the problems of 
psychology on these borderlines. What I did was to apply the same 
logic to the sociologist that I had applied earlier to the economist. 
Only much later has this logic been systematically applied to the 
cultural borderlines of social systems.^ 

At one stage of my career the question of entering fully into em- 
pirical research in a technical sense naturally arose. The study of 
medical practice and later that of social mobility went a certain dis- 
tance in that direction. I have the feeling that some of the critics 
represented in this volume, and many more elsewhere in the pro- 
fession, have felt that the abandonment of this possibility has been 
a fatal one, because of the importance of the task, if theory is to be 
fruitful, of making it "operational" in the detailed research sense. 
However this may be, my own course has been a diflFerent one. I 
hope to be believed in expressing the deepest respect for competent 
empirical research and the conviction of its central importance, in- 
deed utter indispens ability in the building of a science. However, at 
the same time I wish to contend for the justification of specializa- 
tion in theory. If one is to be a specialist, his concern with em- 
pirical materials may justifiably be couched in terms of considera- 
tion of their theoretical significance, and with their codification in 
relation to such problems rather than their original production. It 

3 Cf. "Culture and the Social System," Introduction to Part IV of 
Theories of Society, Talcott Parsons, Edward Shils, Kaspar Naegele, & 
Jesse Pitts, (eds.), Glencoe, III: The Free Press, 1961. 


is my present conviction that, even apart from matters of tempera- 
ment and of serious gaps in research training, commitment to major 
programs of empirical research in the usual sense would have been 
incompatible with the following through at the same time of a 
major program in the building of general theory. I see this whole 
problem as one of the differentiation of function in a complex sys- 
tem of culturally oriented interaction. 

In any case, my empirical interests have been highly diversified, 
and such contributions as I have made have been mainly at the 
level of summary and interpretive essays rather than "research." 
I should, however, maintain that such contribution can be gen- 
uinely empirical, since there is no clear line between "hard data" 
and more general statements of fact which can be worked out with- 
out or only partially involving technical research procedures, partly 
by using the technical findings of others more or less directly, partly 
by putting together evidence from a variety of sources. This point is 
related to a certain pattern of "occasionalism" in writing empirically 
oriented essays. I have tended to be pragmatic in hoping to find 
interesting theoretical applications and implications in a variety of 
the problems posed at meetings, in printed symposiums and the 
like. In recent years, the emphasis in these connections has tended to 
center on various facets of American or contemporary industrial 
society and have expressed as a primary theoretical interest the 
analysis of the large-scale social system as a system."* 

In the framework of this strategy, if it deserves such a name, it 
has seemed natural that, from time to time, attempts should be 
made to reach higher levels of more generalized theoretical codifi- 
cation and statement. These have included the first fragmentary 
promulgation of the pattern variable scheme (1939), the codifica- 
tion project which eventuated in Toward a General Theory of Ac- 
tion ( 1951 ) , with its generalization of this scheme and the placing 
of it in the setting of a more general statement of the action frame 
of reference; the collaboration with Bales and Shils on the Working 
Papers (1953) with its double focus on functional "system prob- 
lems" and phases of process in time, the development of the sug- 

* Cj. my recent volume, Structure and Process in Modern Societies, 
Glencoe, 111.: The Free Press, 1960. 

320 • Talcott Parsons 

gestions made there oiF an input-output schema in Economy and 
Society, and the very recent attempt at a more systematic exten- 
sion and reformulation o£ the pattern variable scheme.^ 

In a certain sort of retrospect, it seems to me that each of these 
attempts has, in strictly theoretical terms, represented an important 
advance. It has not, however, been in any simple sense a "linear" 
advance, but has fallen in the pragmatic tradition in taking ad- 
vantage of opportunities to clarify and extend analysis in relatively 
particular theoretical fields and to articulate the new material with 
the developing general scheme. Articulation in this sense has of 
course been a two-way process; there has been extension of the 
general scheme into new problem areas, but also and necessarily 
modifications of the previous formulations of the general scheme 
and its applications in the light of the newly emerging considera- 
tions. Since this has happened in a pragmatic way, it has naturally 
been a source of confusion to people trying to follow the develop- 
ment who have not been intimately involved with the particular 
phases under intensive consideration at the moment. 

It is perhaps in the nature of the type of pragmatic development 
which has been sketched here, both that it should be, as a theoreti- 
cal enterprise, the generator of substantial amount of resistance 
even to its scientifically meritorious, features, and that it should be to 
a peculiar degree thrown on the critical judgments of the relevant 
professional groups— the latter point is relevant particularly because 
of abandonment of the protection of a rigid ideological position 
which has figured so prominently in Continental Europe. It is at 
least tempting to think that this situation may have something to do 
with the recurring complaint, for many years now, about my being 
so hard to understand. Having reached what I hope is a certain "age 
of humility" I am not at all prepared to discount entirely the view 
that there are peculiar and unnecessary obscurities in my writings. 
At the same time I can claim to be somewhat sophisticated in the 
sociology of knowledge and hence in tlie interpretation of resist- 
ances to certain types of intellectual innovation. In this role I can- 
not entirely dismiss the possibility that some of the complaints may 

^ "Pattern Variables Revisited," American Sociological Review, Au- 
gust, 1960. 


be manifestations of such resistances. In any case, it is not possible 
for an author to be fully objective about the reception of his work; 
any more ultimate judgment will have to be left to the outcome of 
the process of natural selection through professional criticism by 
which scientific reputations ultimately come to be stabilized. 

In more general terms, I think that in this section I am attempting 
to present an interpretation of the nature of the enterprise which is 
the "general theory of action" as an alternative to the critical ap- 
proach of Professor Black. Notwithstanding some statements which 
I have made on occasion, my present considered opinion is that, 
though it has moved in that direction, my approach is not yet a 
logico-deductive system, but rather a temporal and historical series 
of contributions toward the development of such a system. Above 
all I would reject the rigid alternative: either a fully integrated de- 
ductive system or a congeries of unrelated conceptualizations and 
generalizations. I should contend strenuously that the level of the 
Structure of Social Action represented genuine systematization, at a 
certain rather elementary level, to be sure, but well in advance of 
previous attempts. The steps taken since then have by and large 
been real advances from that point, advances by extension, but also 
clarity of definition, analytical refinement, and better theoretical 

Perhaps, with due caution, it is permissible to introduce an anal- 
ogy between the process of theory-building in a developing scien- 
tific field and the process of development of a legal system. There 
is a sense in which more general theory is to a field of science what 
the more general legal principles are to a legal system. Many legal 
philosophers have of course thought of the ideal legal system as 
one for which most even quite detailed precepts could be logically 
deduced from first principles. One may doubt whether any visible 
system of law has ever concretely been developed in this way, cer- 
tainly least of all Anglo-American Common Law, which has been 
built up bit by bit from cases, gradually widening its ranges of 
generalization. Here the function of the theorist in science may be 
Hkened to that of the appellate judge whose primary function for 
the system is not the disposal of cases, but rather the interpretation 
of rules at the higher levels of generality, their codification in rela- 
tion to general principles, testing for consistency and the like. 

322 • Talcott Parsons 

Such a process could not be fruitful if it were purely eclectic, as 
some of the "legal realists" have contended was the case. There has 
had to be a relatively determinate fundamental orientation; in the 
case of Common Law I think this can be said to have been attained 
in the 16th and 17th centuries. But within this orientation, I think 
legal development has been the kind of pragmatic process I have 
been outlining. 

It is my judgment that the element of pragmatism is more im- 
portant in new than in old sciences, and in sciences like those of 
behavior where the definition of the theoretically significant varia- 
bles ( at many different system-subsystem levels ) presents very great 
difficulties, where system-reference problems are peculiarly com- 
plex and where, probably, the play of extrascientific considerations 
into the process of science-building is a more serious source of di- 
version and confusion than in the physical sciences. But precisely 
for reasons of this sort, what I have just called a relatively deter- 
minate fundamental orientation is of the first importance. 

It would be my contention, as noted above, that for the sciences 
of action the outline of this fundamental orientation was in fact 
present in the main Western intellectual traditions by the late 19th 
century; that its most fundamental component came from the utili- 
tarian tradition, but that a contribution from the more "collectiv- 
istic" sources of French "Rousseauism" and German Idealism was 
necessary.^ This general orientation came to focus in the converg- 
ence which was documented in my Structure of Social Action. The 
authors treated there in fact did a great deal of theory-building in 
the more specific sense, but, to me, the great historic event of con- 
vergence opened the way to a much more detailed and technical 
phase of the process which has been going on apace in the genera- 
tion since their work was completed. 

The work on general theory on which I have been engaged seems 
to me to He at the upper part of the pyramid of levels of generality 
which the theoretical structures of a developing tradition such as 
this must comprise. As such it interpenetrates at many points with 
Merton's preferred "middle range" level where such things as ref- 

• Cf. "The General Interpretation of Action," Introduction to Part I, 
Section A, in Theories of Society. 


erence group theory fall. At a still lower level (in terms of logical 
generality, not of general scientific importance) lie the more tech- 
nically "operational" problems to which Williams refers. All of 
these (and more refined distinctions could of course be made) are 
essential ingredients of a developing science. None is the simple 
prerequisite of the others, but all typically are developing concur- 
rently. Necessarily, in the course of the development, serious im- 
perfections in their coordination appear, which require difficult 
critical work to be ironed out. The present symposium seems to me, 
among other things, to be an important contribution to this essen- 
tial task. 

Let us now turn to some more substantive considerations. The 
fundamental orientation to which I have referred must, if it is to 
serve as such, have a certain relative stability; it is perhaps a kind 
of unwritten constitution of the scientific field. This does not, how- 
ever, mean that the rigor and consistency of its formulation is not 
subject to improvement. Professor Black has, I think, provided an 
attempt at such a formulation in the list of eight "assumptions" of 
general action theory which he reviews in his paper (pp. 272-73). 
This formulation provides a convenient point of reference for a few 
considerations at this level. To save space, let me simply list them 
here in abbreviated form: 1) All action is directed toward goals; 
2) all action is relational; 3) all human response to stimuli has the 
two dimensions of 'cognitive' and 'cathectic'; 4) all action involves 
selection among alternatives; 5) selection involves the use of stand- 
ards; 6) all interaction involves complimentarity of expectations; 7) 
orientations and actions are organized in systems; and 8) the above 
principles apply to social systems at all levels of complexity. 

This is Black's version of what I would call the "frame of refer- 
ence" of action theory. I myself have on different occasions put for- 
ward a number of different formulations, none of which exactly 
coincides with his. It would lead too far afield to attempt here 
critically to codify these with each other and with his; what T 
should like to attempt is. rather, a new succinct statement which 

324 • Talcott Parsons 

the reader can compare with Black's. This statement is made pos- 
sible by the new developments reported in the paper referred to 
above ("Pattern Variables Revisited," op. cit.). 

Perhaps the most ultimate principle may be said to be that o£ 
duality, which is perhaps phrased in Black's item of relational qual- 
ity (2) but also relates to his assumption (3). The primary state- 
ment of this concerns the relation between actor and situation; one 
cannot speak of action except as a relation between both, it is not a 
"property" of one or the other or of the two as aggregated rather 
than related. You cannot have a relation without a minimum of 
two terms to it. (Comparable cases are the subject and object of 
epistemology, or the pair concept heredity and environment. ) 

Second, the relatedness of pairs of relata is spelled out in two 
primary directions. One is that of normative control (in the cyber- 
netic sense), or the control by a more highly organized entity over 
one which is less highly organized, which stands in a "conditional" 
relation to the former. This is the internal-external dimension of 
spelling out. The other is that of the temporal process of imple- 
mentation of "need" or "pattern" (whatever term is used), that is, 
the transition of state in time from "potentially" to "actuality" (or 
the frustration of such a transition). Here the duality concerns on 
the one hand the elements in which the continuity of properties 
entailing potentiality is conceptualized, on the other hand the re- 
sponsiveness of the actor to the immediacy of situational exigencies, 
looked at both as dangers and as opportunities. 

Underlying both of these is the conception of the relevance of the 
cultural level of categorization in terms of meanings. This implies 
that an essential point of reference must be a postulated "knowing" 
and "feeling" (Black's assumption, paragraph 3) unit of reference, 
an "actor" for whom the objects of his situation have meaning. This 
is the famous Weberian "subjective" point of view (Verstehen) 
which, as Devereux rightly points out, has always been essential to 
the scheme. 

Though particular orientations of isolated actors to situations may 
conceivably occur, this is a limiting case of secondary theoretical 
interest. The theoretically general case is that of plural actors, 
interacting with each other so that each concrete actor-unit be- 
comes situation to the other; in the terminology I have used, each 


unit is both an actor and a social object. These are thus not cate- 
gories of concrete entities, but analytically distinguishable aspects 
of the same entities. This is the concept of system (Black, 7) which 
is of direct interest to the theory of action; that is, a system is seen 
as two or more interacting units which are at the same time actors 
and social objects to each other. 

Given the postulate of the hierarchy of normative control it fol- 
lows from this that as a condition of minimum integration of such 
a system there must be some degree of complementarity of ex- 
pectations (Black, 6). The alternatives are randomness of orienta- 
tions relative to each other, or a level of direct conflict which would 
not be compatible with the continuance of such a system. The con- 
cept of boundary-maintenance as a criterion of such a system also 
follows from the combination of normative control and mutuality 
of orientation. 

Every such system must by definition have an environment which 
is external to it, vis-a-vis which there is a boundary— which may be 
complex— and relative to which there is a problem of "control," i.e., 
of maintenance of the pattern of the system vis-a-vis the fluctuating 
features of the environment. The inherent possibility of plural sys- 
tems, and their differentiation into subsystems, seems to belong to 
the general logic of science. But that at least two such levels must 
be involved in the analysis of action, is among other things a con- 
sequence of what I have called the subjective point of view. This 
is to say, the observer of a system of action, as scientist, must him- 
self in some sense be conceived of as an actor. But the system he 
observes, or its units, e.g., individual persons or organized collectivi- 
ties of them, must also consist of actors which is to say they belong 
to the same general category of objects which includes scientific 
observers. There is therefore a sense in which in the action field the 
act of scientific observation is a process of action in interaction with 
the objects observed, that, therefore, observer and observed taken 
together constitute a system of action. If, for example, there were 
no "common culture" in this system, there would be no way of 
"interpreting" what the acts of the observed meant to the actor 
within his system. Therefore, in some sense, the system observed 
must be a subsystem of a larger one of which the observer-observed 
relation is a part. Hence Black's paragraph 8 is a necessary part of 

326 • Talcott Parsons 

the scheme on some level (though other subsystems of action than 
the social may be involved). 

The remaining three of Black's assumptions (1,4,5) seem to me 
to be direct consequences of the two first-order spellings-out of the 
duality principle as combined with the cultural reference. Norma- 
tive control clearly means that the relation of higher-order systems 
to lower-order conditional systems must be selective; random ac- 
cessibility to situational-environmental influence would be incom- 
patible with the imperative of maintenance of organization, i.e., with 
order. If the subjective postulate is accepted, it means that the 
mechanism by which such selectivity operates must involve a com- 
ponent normatively meaningful to the actors in the system as well 
as to the observer; a standard may be conceived of as a selective 
principle which has normative meaning to a relevant actor. Finally, 
in its relation to the external situation, an action system must be 
directed in some sense toward "optimizing" the relation between its 
internal "needs" and the significant features of the situation. Be- 
cause, however, of the changing nature of environments (relative 
to action system), an optimal relation is necessarily limited in scope 
and in time; it is as such a relatively optimal segmental state that 
I should conceive a goal-state to be, and the property of goal- 
direction as the tendency to act in the direction of attaining such 
states. It is an essential, though not, I think, the most fundamental 
property of action systems. I should not put it first in my own list 
of assumptions. 

Perhaps one more assumption, not included in Black's list, is es- 
sential, namely that action theory is concerned with the analysis 
of aspects of the behavior of living organism; particularly that 
phase which involves the control and direction of such behavior 
through culture-level symbolic systems and the organization through 
which that control is implemented. There are two points at 
which this assumption becomes essential. The first is that it es- 
tablishes basic continuity with the biological world. Action is es- 
sentially a level of organization of the phenomena of life which 
can be presumed to have emerged in the course of evolution {cf. 
Scientific American, September 1960, especially the article by Hock- 
ett on language). The second point is to draw the line vis-a-vis 
physical behavior. This is not, as such, action in the analytical sense, 


but is controlled by action processes. Thus, I should answer Whyte's 
complaint that "there is no action in action theory" by saying that he 
attempts to find this kind of action at the wrong level in the total 
organization of living behavioral systems. He means essentially the 
physical behavior of organisms. 

This assumption also underlies the very central point which does 
not figure in Black's discussion, that there is a plural hierarchy of 
subsystems of action, other than within social systems, namely the 
behavioral organism, the personality of the individual, the social 
system generated by interaction, and the cultural system organized 
about patterns of meaning.'^ This set is arranged as a hierarchy of 
control in the above sense, from lower to higher in the order stated. 

Within this frame of reference it is possible to say certain gen- 
eral things about the nature of systems of action. The first is that 
the notion of a hierarchically ordered ( in the control sense ) bound- 
ary-maintaining system implies the notion of function, as operation 
relative to a set of exigencies, namely sets of conditions, internal 
and external to the system, which can be shown to set limits to var- 
iation which is compatible with the integrity and effectiveness of 
the system. There are ranges of tolerance, but beyond these, proc- 
esses of fundamental change, including dissolution, will be set in 
motion. The concept of function used here is essentially the same 
as that used in the biological sciences, e.g., as expressed by W. B. 

Though for particular purposes a much longer list may be needed, 
for the most general theoretical purposes it has turned out that a fist 
of only four is adequate, the four which are generated by the two 
dimensions of spelling out of the fundamental relational duality of 
action discussed above. We may put it that the most elementary 
notion of action implies two functional references, namely, ( 1 ) the 
maintenance of a pattern of orientation and ( 2 ) the definition of the 
significance or meaning of one or more situational objects. The es- 

'' This schema of the four basic subsystems of action is most fully 

developed in the essay "An Approach to Psychological Theory in 

Terms of the Theory of Action," in Psychology; a Science, Sigmund 

Koch (ed.), Vol. III. Cf. also Introduction to Part IV, Theories of 


8 W. B. Cannon, The Wisdom of the Body, New York: Norton, 1932. 

328 • Talcott Parsons 

sential point is that though by definition always both are involved, 
these are analytically distinct and not reducible to each other; their 
reduction would imply, either, as in much "idealistic" thought, that 
situations were not independent of the orientations of action, but 
were "emanations" of them, or conversely "materialistic" orienta- 
tions would be an epiphenomena of situations. These are referred 
to as the "orientation" aspect of function and the "modality" aspect, 
respectively ("Pattern Variables Revisited," op. cit.). 

When we have a system composed of two or more interacting 
actors in an environment, however, two further exigencies have to 
be taken into account which are not reducible to either the com- 
ponents of the elementary pair just discussed or to their combina- 
tion. These are ( 1 ) internally, the exigencies of stable interrelation 
of the action units vis-a-vis each other, and (2) the meeting of the 
exigencies of generalized relation between its system and the ex- 
ternal situation conceived as a set of facilities and conditions of 
operation of the system. These are the functional References of in- 
tegration and adaptation respectively. For a system, we contend, 
this is a minimum set; the operation of a system cannot be ade- 
quately analyzed in terms of less than four mutually independent 
ranges of variation for purposes of general theory though of course 
particular problems may be treated in analytically simpler terms. 

This functional schema may then be applied in connection with 
the general distinction characteristic of all scientific treatment of 
systems between structure and process, concepts which I do not 
think need to be explicated here. Each of these in turn can be di- 
vided into relational and unit categories. In the case of structure, 
units in action systems are on the one hand actors, on the other 
hand social objects; the most familiar case is persons-in-roles as 
units in the structure of social systems, but in more complex social 
systems collectivities are also units, as can also be complexes of 
norms and values.^ Relational components of structure then are 
those comprising the stable elements in the relations between units. 
It follows from the normative control aspect of action systems that 
these are in some sense definitions of the "right" or "proper" rela- 

^ The term "unit" is here used in the sense in which a particle is a 
unit in a physical system, or a cell in an organism, not that in which 
a unit of measurement may be referred to. 


tionship pattern. They are clearly of primarily integrative signifi- 
cance in a functional sense. 

When we come to process, the components which are parallel to 
units in a structural sense are categories of input and output, ac- 
cording to the level of system reference conceived either as oper- 
ating between the system itself and its environment or as between 
subsystems (units) in relation to each other. Thus, at one level 
process may be conceived as the "passing" or exchanging of inputs 
and outputs between systems or subsystems. Because of the sym- 
bolic-cultural aspect of action, it may be said that the fundamental 
input-output categories are all "informational" and that the basic 
action process is always in some sense "communication." 

Very simple input-output interchanges may be direct, like barter 
in the economic case. But in complex systems generalized media 
become necessary, media which on the one hand are rooted in the 
structure of the system, on the other hand can "circulate" as regula- 
tors of the interchange process. These media define relations in the 
context of process. On the most general human level language is 
the prototype of such media; the best understood one within com- 
plex social systems is money. It seems probable that pleasure is 
such a medium in the behavioral organism in its relation to the 

It has been contended that within the normative-control and the 
subjective-cultural features of the action frame of reference the 
principle of relational duality is of fundamental significance. It 
seems to me that it is here that the pattern variables fit. They consti- 
tute the minimum differentiated spellings out of the fundamental 
duality which, in connection with the functional principles just out- 
lined, are necessary to define the essential functional problems and 
hence conditions of stability, of a system of action. This involves 
the questions, raised by Professor Black, of the stability and ex- 
haustiveness of the list, as well as of the meaning of their dichot- 
omous structure. 

With respect to number, I should say it must be either four or 
six pairs and the choice is a matter of definition. The basic four, 
which in turn are divided into the "orientation set" of specificity— 
diffuseness, and affectivity— neutrality, and the "modality set" of 
universalism— particularism, and performance— quality, are clearly 

330 • Talcott Parsons 

modes of classifying the basic structural and processual components 
of action systems referred to above. Here (as shown in Figure 12, 
taken from "Pattern Variables Revisited," American Sociological 
Review, Vol. 25, August 1960) they serve to make the elementary 
distinctions, a) between the two sides of the fundamental orien- 
tation duality, namely, orientations of actors and modalities of 
objects in the situation and b) between the four basic functional 
meanings, which are generated by the cross-classification of each 
of the two parts within each set. Then for defining the relational 
components of the system, that is, integrative standards on the 
one hand, adaptive media on the other, two sets of cross-classifica- 
tion are used, with pattern variable components drawn from each 
of the two elementary sets. 

As I have shown in the paper referred to, within these assump- 
tions and at this level the set of four pattern variables and their 
classificatory combinations are in fact exhaustive. In my present 
opinion their status is not either arbitrary or independent of the 
concept of system under the general frame of reference just re- 
viewed. They constitute an integral aspect of the system of action 
as that has been in process of crystallization. 

This means that of the five, as published in 1951, one, namely, the 
self-vs. -collectivity orientation variable, is a special case. As was 
clearly recognized at that time it does not belong to either of the 
two elementary sets, but stands on a more general logical level. 
As I now see it, it is the formulation in one special case of one of 
the two general spellings out of the original duality, the more gen- 
eral statement of which has appeared in recent writings as the 
external-internal dichotomy. If it is to be called a pattern variable 
at all, then it must be matched by a second one which above I have 
referred to as the "instrumental-consummatory" distinction. The ele- 
mentary sets, as it now seems to me, define the components of a 
system of action and those of their combinations which state the 
basic interrelations of those components; the supplementary pairs 
do not do this, but rather they define the axes of differentiation of a 
system of action in relation to the environment external to it. It 
seems to me essential to keep these two levels of categorization 
distinct and for this reason my present inclination is to reserve the 
term "pattern variable" for the categories which classify components 

( Adaptation) 


{ Goal- Attainment) 

Adaptive exigencies represented by 
'Symbolic' Meanings of Objects 



— >— Univ 
1 Diff 





— >-Qual 
i Aff 







o Universalistic 







Modalities of Objects 











Orientations to Objects 

Integrative Standards for Orientation 








t Univ 
-< — Spec 













I Part 
-< — Diff 



( Pattern-Maintenance ) 

Figure 12. The components of action systems. 

332 • Talcott Parsons 

and use another term to designate the axes of diflFerentiation, in- 
deed that term itself may well be appropriate. 

The classifications which are introduced within each set of pat- 
tern variables are arranged in terms of the scheme of four basic 
functional exigencies of systems which has just been reviewed. The 
orientation set then characterizes a unit of an action system treated 
as actor whereas the modality set characterizes a unit treated as 
object; it should be remembered that in action systems any con- 
crete unit may be treated as both: the distinction is analytical. The 
treatment of each of these aspects of the unit in terms of four dif- 
ferent possible combinations of elementary components even within 
the set is necessitated by the fact that every unit of a system of 
action may, by shifting the level of system reference, be treated as 
itself a system, hence as subject to all the exigencies relevant to any 
system of this class. The attribution of the properties of a system to 
any given level within a macroscopic-microscopic, or system-sub- 
system range is a matter of the particular empirical problem-state- 
ment and is not ontological in significance (this point was strongly 
stressed in Working Papers). 

Seen in this context the dichotomous logical structure of the pat- 
tern variable scheme is, as noted, a consequence of the basic dual- 
ity of the frame of reference as spelled out in the context of the 
concept of systems of action at not one but a minimum of two 
system-subsystem levels. The formulation actor-situation as a state- 
ment of the basic duality may in one context be regarded as a log- 
ically general one which, in terms of the breakdown of systems of 
action into four major types of subsystem, has the following four 
more special forms: namely, organism-environment (for behavioral 
organism); actor-situation in a more specific sense (for personality 
system as actor); system-environment, again in a more specific 
sense (for social system as actor); subject-object (for cultural sys- 
tem as reference ^^ ) . 

There is an obvious terminological diflSculty here, one of a very 
common sort in the history of science. This is that a concept, or the 

^° Cf. Talcott Parsons, "Some Reflections on the Problem of Psycho- 
somatic Relationships in Health and Illness," forthcoming paper in 
Psychosomatic Medicine. 


concept pair embodying a logical distinction, has been used for at 
least two diflFerent levels of logical generality in relation to a more 
general frame of reference. The process of logical differentiation of 
the theoretical scheme has then made it necessary to distinguish 
the levels, whereas that had not previously been necessary. Often in 
such a case there are no terms available in current usage to state 
the double distinction, or the adaptation of such terms does violence 
to many implicit understandings in current usage. Unfortunately I 
have no ready solution of the present terminological difficulty, but 
I hope the logical situation is clear.^^ 

Returning, however, to the most general level, we may use the 
concepts actor and situation to designate the most basic duality of 
the scheme. In pattern variable terms this takes the form of the 
distinction between the two sets of pattern variables. The next step 
in the "derivation" of a system in these terms is the introduction of 
the distinction, on both "sides" of the relational scheme, between 
the two directions of "spelling-out" discussed above, namely ( 1 ) the 
hierarchy of control (internal-external) and, (2) the process of im- 
plementation in time ( instrumental-consummatory ) . This yields the 
double fourfold table scheme in terms of which each of the two 
elementary pattern variable sets is arranged as differentiated in 
terms of the general functional frame of reference, and the two 
sets in turn are treated as constituting the "pattern-maintenance" 
and the "goal-attainment" references respectively. Thus, the pattern 
variables are directly built into the minimum conceptuahzation of 
the analysis of any process of action. 

11 The same kind of difficulty has appeared at various other points in 
this field. Thus, an illustration very close to the present dilemma is the 
problem presented by the conflict over the uses of the terms "culture" 
and "social system"; broadly much American anthropological tradition 
has meant by culture something which has included what sociologists, 
e.g., in the Durkheimian tradition, have meant by social system. Some 
way of designating the analytical distinction has now become neces- 
sary (cf. A. L. Kroeber & Talcott Parsons, "The Concepts of Culture 
and of Social Systems," American Sociological Review, October 1958). 
A similar dilemma has arisen with reference to the formula, the 
"behavior of the organism," which is still the preferred one among 
most American psychologists. This fails to distinguish the levels I have 
spoken of as that of the behavioral organism on the one hand, the 
personality on the other. 

tf o i2 

tfl w » 

^ ^8 


"i 6 

O S 

o B 




S ^ 

1-1 Oi 

° 5 

?> S ■< 

o a 

o .?d. 

o c ^ 


i2 o H, 

CO < 











a -2 



















Ground -of- meaning 


Manifold of Part 
evaluative Diff 


Allocative Selection 

Range of Aff 

GOAL (attainment) 

Empirical- Univ 
cognitive Spec 












• suo^ipuoo Suniran JO uoi;o9jT(i 

I© j^uoo JO uo];o9 Jia . 


When, however, we take the step to the analysis of systems, as 
discussed above, it becomes necessary to consider the two further 
functional exigencies which have been called integrative and adap- 
tive. These are {cf. "Pattern Variables Revisited") categorized in 
terms of combinations of the pattern variable components which 
draw from both of the two elementary sets, in the one case char- 
acterizing integrative standards (Black, 5) and in the other case 
generalized mechanisms or media of adaptive function, i.e., the 
"representation" of the external environment within the system. 
Within a set of rules, which have been stated in the paper referred 
to, the combinations of the pattern variables involved in these four 
functional references of systems of action are exhaustive and, for a 
logical scheme, they cover what have been claimed to be, at the 
requisite level of logical generality, all the basic functional aspects 
of a system both structural and processual. This latter claim is for- 
mulated in terms of the transformation from the abstractly func- 
tional mode of presentation of the combinations ( Figure 1 ) to the 
structural-processual mode (Figure 13, taken from "Pattern Varia- 
bles Revisited" [p. 476], American Sociological Review, Vol. 25, 
August, 1960). 12 

The above considerations which, it should be made clear, could 
not have been formulated in terms of the state of my thinking in 
1951— the latest materials which Black considers— constitute my an- 
swers to his questions about the nature of the assumptions on which 
the theory of action as a conceptual scheme rests, the nature of the 
pattern variables, whether the list is exhaustive, and whether they 
are "necessary" within the scheme. On the question of the nature of 
the assumptions, I think I have adequately accounted for all of the 
eight propositions stated by Black, but I have placed them in a 
somewhat dijBFerent order and have related them more explicitly 
to each other and to some which Black does not bring out. Of these 

12 The above account of the pattern variables and the way in which 
they fit into the general action scheme is necessarily, in view of 
limitations of space, so condensed that to many readers it may appear 
cryptic. Unfortunately it is simply impossible to give a full exposition 
here without sacrificing too many other things which need to go 
into this essay. Hence the best I can do is to refer the reader who has 
difiiculty to the paper "Pattern Variables Revisited" which is easily 

336 • Talcott Parsons 

the most essential are the basic cultural references of action theory, 
and the special place of the hierarchy of control and the temporal 
implementation themes as the first-order spellings out of the funda- 
mental relational scheme. On the question of the pattern variables, 
I think it can now be said that they are essential ^^ and that they 
are exhaustive, though not so in the 1951 formulation. The link be- 
tween the frame of reference or assumptions and the pattern varia- 
bles is clearly supplied by the conception of systems as subject to 
four functional exigencies, a conception which, though it had been 
formulated in considerable detail as early as 1953 (Working Papers), 
Black did not take into account. 

What, according to eventual terminological decision, may be either 
the sixth pattern variable, or a category on another level (as 
I incline to treat it), namely, the instrumental-consummatory dis- 
tinction interpreted as the basis for the lining of structure and proc- 
ess, constitutes, I think, the basic answer to the objection which 
Black, among many others, raises, alleging a "static" bias to be in- 
herent in the scheme. Here I think it is essential to make the funda- 
mental distinction between the concrete developmental processes by 
which a conceptual scheme is built up and the logic of the scheme 
looked at in more general terms at any given point of its develop- 
ment. There is, in my opinion, more truth in the allegation in the 
first than in the second context. 

It is quite true that the empirical-theoretical problem which was 
at the focus of my own theoretical "take-off" was the problem of the 
bases of social order; it was, as Devereux quite clearly points out, 
the problem posed long ago by Hobbes. It was the fact that Hobbes 
had never been satisfactorily answered within the utilitarian tradi- 
tion that made recourse to the intellectual resources of idealistic- 
coUectivistic traditions, especially by Durkheim and Weber, nec- 
cessary; it is, in my opinion, out of the "marriage" of the two tradi- 
tions that modern sociological theory has been born. It turns out, 
I think, that consideration of the problem of order in these terms 
leads directly into the nature and bases of the structure of action 

12 If they were not used, essentially the same concepts under different 
names would have to be introduced. 


systems, in particular of social systems, and structure is by definition 
a category of the relatively stable aspects of a system. 

Professor Black makes the interesting point that the concept orien- 
tation, w^hich for the personality is comparable to that of attitude, 
is in fact a structural concept and designates a relatively stable 
aspect of a system. In this, in my opinion, he is correct, but not in 
drawing the inference that because of the prominence given to the 
concept of orientation, the theory is in general biased against dy- 
namic analysis. This is simply because, in the general scheme, the 
orientation component is only one of four which are treated as 
equally fundamental aspects of systems. It is quite correct that this 
is treated consistently as the most stable of the four, but it does not 
follow that there is any disposition to discriminate— intellectually, of 
course— against the other three. It does, however, follow that proc- 
esses of change which involves changes in the orientation factors 
( in the sense of Figure 1 above ) are of a different order from those 
involving the other three functional components; but surely it is 
not inherently "bias" to assert that, among a classification of four 
factors, one may have properties not shared with the other three. 

This brings one directly to the equally controversial concept of 
equilibrium. In my opinion, this is overwhelmingly an ideological, 
not a theoretical, question. Theoretically, the concept of equilibrium 
is a simple corollary of that of system, of the interdependence of 
components as interrelated with each other. In turn, the concept 
of system is so fundamental to science that, at levels of high theo- 
retical generality, there can be no science without it. If there are 
no uniformities involved in the interdependence of components 
there is no scientific theory. Furthermore, a fundamental distinc- 
tion always needs to be made between analytically formulated rela- 
tions of variables, what more methodologists mean by laws, and 
empirical interdependences of operationally identifiable compo- 
nents. Equilibrium is, in my opinion, a concept defining the relation 
between the two. It states that, given the concept of an empirical 
system, there are in fact empirical conditions of its stability— 
whether this stability be "static" or a moving stability-in change. 
If no distinction can be made between conditions which favor sta- 
bility and those which tend to a change of state away from the 

338 • Talcott Parsons 

"stable state" in either of these senses, there can be no such thing 
as systematic empirical analysis. 

These extremely general considerations are given added signifi- 
cance for systems defined as involving cybernetic control and the 
closely related conception of boundary-maintenance. Thus, for an 
internal combustion engine, there must be a difference between the 
consequences of a) the existence of a stable relation betv^een the 
position of a "throttle" and the rate of fuel-input in the combustion 
chambers, and b) a situation where the fuel input and changes in 
the position of the throttle are wholly unrelatable; an airplane pilot 
who attempted to use the latter type of throttle as a basic control 
would surely be courting disaster. 

Equilibrium, in short, is nothing but the concept of regularity 
under specific conditions as applied to the internal state of an em- 
pirical system relative to its environment. This regularity of course 
should always be treated as relative rather than absolute; indeed, 
it is generally subject to considerable ranges of tolerance, and of 
course its maintenance is by no means inevitable but, if the condi- 
tions on which it depends are changed beyond certain limits, it will 
disappear, again most probably giving way to other regularities 
than to sheer randomness. Thus in my opinion this concept is an 
inherently essential part of the logic of science, of importance pro- 
portionate to the level of theoretical generality aimed at. The denial 
of its legitimacy in the conceptual armory of social science is at the 
least, in my perhaps not very humble opinion, symptomatic of the 
denial that social science itself is legitimate, or realistically pos- 
sible. On this point I have thus remained completely unimpressed 
by the barrage of persistent criticism. 

It should be clearly understood that not only are equilibrating 
processes very frequently doubtful in their outcome so that break- 
down of equilibrium is scientifically as important a phenomenon as 
its preservation (but of course not inherently more important), but 
also that equilibrium itself is neither attained nor maintained simply 
by the persistence of some "static" factor. That something, e.g., a 
"pattern of orientation," should remain unchanged, seems to me to 
be a necessary component in a state of equilibrium, but it is 
equally essential, in the light of presumptive change in the environ- 
ment of the system, that some things change as a condition of main- 


tenance o£ equilibrium. This is completely crucial in the biological 
theory referred to; thus, in order for the mammalian organism to 
maintain a relatively constant body temperature, there must ob- 
viously be some trend of change of environmental temperature. If 
nothing in the human organism changed, between an environ- 
mental temperature of 110 degrees Fahrenheit and 10 degrees- 
discounting the eflFect of clothing— it would obviously be impossible 
to maintain constancy of internal temperature. This is the whole 
point of Cannon's famous concept of homeostasis, which I take to 
be a special case of that of equilibrium. To say that such a pattern 
of analysis is simply a manifestation of a "static bias" seems to me 
to be a gross distortion. 

This point provides a convenient transition to two final questions 
raised by Black, on which I should like to comment briefly. These 
are, namely, the questions of the extent to which the assumptions 
of the theory of action are grounded in solid empirical generaliza- 
tions, and of whether or not the scheme gives rise to important 
generalized hypotheses (or theorems). 

I have stated the relation between the concept of equilibrium 
and that of order. Whatever its Umits and however precarious it 
may be, order in empirical action systems is a fact; that it is problem- 
atical is proved by the commonness of its (relative) absence. Society 
is not, in fact, in general a Hobbesian war of all against all; this 
fact was the major focus of the great theories of order, particularly 
Durkheim. But social disorder certainly exists; witness the condi- 
tion of the ex-Belgian Congo in the late summer of 1960. Order, this 
is to say, is a very real phenomenon, but also is problematical, as 
Devereux and Williams correctly point out. 

That order in systems of action is grounded in normative control 
is, it seems to me, a very basic empirical generalization which has 
become increasingly substantiated and clarified in the past genera- 
tion, above all by developments in the area between the biological 
and the social sciences. The older "mechanistic" theories have been 
pretty thoroughly discredited, as far back as the theory of control 
of physical systems. Cybernetic and information theory at this very 
general level link up with the kind of thing represented in behavior 
psychology by Tolman in speaking of "purposive behavior in ani- 
mals and men," and more recently by work on the functioning of 

340 • Talcott Parsons 

the brain ( cf. above all Olds and Pribram ) . Black's statement of 
the assumption of goal-directedness is one special case of this 
broad generalization, but the role of selectivity and of standards 
also belongs in this context. If there is any major empirical general- 
ization embodied in the trend of scientific thinking with reference 
to the "life sciences" in our time, the imputation of normatively 
controlled order to living systems, and the postulation of a hierarchy 
of such levels of order, seems to me to be one of the most funda- 

Next, I think it can be quite sharply stated that the scientific 
status of the importance of what is variously called "symbolic 
process," "communication" and a number of other related things, 
has been strongly vindicated. Not least of the relevant considera- 
tions here has been the development of the science of linguistics; 
unless communication means something, surely language is not 
even a phenomenon worth studying. This leads directly to the 
empirical status of the assumption, unfortunately not made explicit 
by Black, that action theory is fundamentally oriented to the prob- 
lems of meaning in the symbolic-cultural sense. 

Such matters as complementarity of expectations are somewhat 
more specialized, but to anyone deeply immersed in contemporary 
social science it is diflBcult to see how the status of empirical 
generalization can be denied to such an "assumption." For example, 
the whole argument of Baldwin's and Bronfenbrenner's papers, 
regardless of how far they agree or disagree with me, would fall 
to the ground. Certainly one of the most important developments of 
social science in the last generation has been that of role theory, 
and for this complementarity of expectations has been fundamental. 
Of course one possible misunderstanding must be forestalled, namely, 
that complementarity should be considered an open and shut 
phenomenon. On the contrary it is a special case of equilibrium; 
relatively complete complementarity of expectations is one major 
condition of the stability of interaction processes. Whether or not 
it in fact materializes is an empirical question in the particular 
case. But if it were not empirically common there would be no 
social systems or personalities in the human sense. 

Broadly speaking, I take it that Professor Black expresses an 
attitude of skepticism toward the empirical status of the assump- 


tions of action theory. In my opinion this would not be possible 
for a man deeply immersed in the recent developments of social 
science. Many of these points have been historically controversial. 
But the radical denial of any of the basic ones I have just reviewed, 
seems to me to be out of court in the present state of the relevant 
disciplines. The position of the old fashioned behaviorist, for exam- 
ple, who denies the scientific legitimacy of "subjective" data, or of 
the mechanist who denies that any sort of "normative" control ever 
operates in the empirical world, are no longer strongly held. Indeed, 
in my opinion these questions are no longer controversial in any 
authentic sense. Of course, from the legitimacy of these assumptions 
on the most general level it does not follow that the much more 
specific ways in which I have built them into a relatively detailed 
conceptual scheme must be accepted; that involves several further 

The conceptual scheme has, in my opinion, now reached a stage 
of development where the principal diflBculty is not in deriving 
generalized hypotheses, but in stating them at the level of generality 
and in the system-reference which is most meaningful for the 
purposes in hand; the complexity of the scheme is such that this 
presents so very formidable a problem that it is out of the question 
to review it in detail here. Fortunately Black expresses very clearly 
the point that a general theory should not be expected to give 
directly the answers of empirical use to the problems of operational 
criteria in particular situations. There must be many sets of such 
criteria for the many different kinds of uses and it is my opinion, 
already stated, that it is too big a task for the same person to be 
the kind of general theorist I have attempted to be and at the same 
time to supply the answers to the relevant operational questions 
over any very large part of the range for which the scheme is 

Here I should like to attempt to state a few hypotheses in the 
most general terms and then to give a few illustrations at the more 
concrete levels of particular types of system. Most of the terms 
used will have been defined (or used) in the preceding outline, 
although I hope the reader will not hold me to standards of the 
fullest technical rigor, but will consider rather the question of the 
meaningfulness of the propositions within the system, and the 

342 • Talcott Parsons 

possibility of working out fully rigorous statements and derivations.^* 
The most fundamental theorem of the theory of action seems to 
me to be that the structure of systems of action consists in institu- 
tionahzed (in social and cultural systems) and/or internahzed (in 
personalities and organisms) patterns of cultural meaning. That 
this is not a proposition obvious to common sense is attested by 
the long and complex history of behavioristic and other reductionist 
theories of human behavior; take, for example, the very recent 
imputations to Freud ( illegitimately, I think ) of a strictly biological 
instinct-determinism theory. 

As indicated in the above formulation the relevant cultural 
pattern-components must be differently formulated for different 
subsystems of action, viz. : for social systems they are values, norms, 
goals of collectivities, and patterns of role-expectation for indi- 
viduals; for cultural systems they are patterns of the grounding of 
meanings, of evaluation, of expressive symbolization, and of empiri- 
cal cognitive ordering of experience; for personality systems they 
are internalized (broadly in the Freudian sense) value-patterns, 
social objects and motivational orientations ("need-dispositions"); 
finally for the behavioral organism they are learned patterns of the 
orientation of behavior, stored in memory. From this point of view 
all cases of learning which result in organized "cognitive maps" 
( Tolman's term ) are in the present sense cases of internalization. 

There are, then, two further basic propositions which state the 
primary conditions of relatively stable institutionalization or inter- 
nalization, as the case may be. The first of these has, for the 
institutional case, been stated as the theorem of the "institutional 
integration of motivation" {Social System, Chap. II). This is that 
the goals of the units of a social system, in an important sense in 
the "last analysis" individual personalities in roles, must broadly 
have the meaning of being contributions to the functioning of the 
social or cultural system of which the units are parts. Like other 

1* The approach used here is somewhat diflFerent from that taken at 
the end of "Pattern Variables Revisited." The cake of theory can be 
cut in different ways to derive hypotheses and this time it seemed 
useful to attempt it through the input-output interchange paradigm 
wliich has not yet been pubhshed in full, but is most fully stated in 
Economy and Society, T. Parsons and Neil J. Smelser, London: Rout- 
ledge & Kegan Paul, also Glencoe, 111.: The Free Press, 1956. 


such propositions this is not one of absolute yes or no in empirical 
reference. All social systems can tolerate a certain amount of con- 
flict and of alienation from normative expectations, expressed as 
rebellion or as withdrawal, but there are relatively definite limits 
which are compatible with the stability of the system, including 
orderly processes of change. The limits in each particular case are a 
matter for empirical determination, but the proposition not only 
asserts the meaningfulness of the question of this determination, 
but also the probability that a relatively high degree of such motiva- 
tional integration will be found to be functionally necessary and 
empirically prevalent. 

For the "individual," as organism and/or personality, the corre- 
sponding proposition concerns the integration of particular needs 
or "motives" with the necessities of maintenance of the major 
orientation patterns of the behavioral system, whether these be 
formulated as "life-goals," "values," or "character type." 

In terms of our more general theoretical scheme this theorem is 
concerned with the relation between the orientation and the 
modality aspect, under the assumption, given in the concept of 
structure, that the latter will be subject to environmental pressures 
(to gratification or frustration) which do not apply in the same 
measure to the former. 

This may, following Durkheim, for the social system case be 
called the "law of mechanical solidarity." It means that, even on 
the most elementary levels, the exigencies of maintenance of a 
structural pattern and the exigencies of "coping" with a changing 
situation must somehow be balanced and hence that a rigid pre- 
dominance of either pressure is incompatible with the functioning 
of the system. 

In proportion as a system becomes complex, however, and hence 
also differentiated, this will not be sufficient. ^^ With differentiation 
of a system into kinds of units, which contribute differentially to 

15 Aggregates of many segmented units which are not differentiated 
from each other are of course theoretically conceivable, but it is not 
difficult to show that such aggregates are hkely to be relatively 
unstable and that, certainly under pressure of such factors as 
"growth," there is an inherent tendency toward increasing differen- 

344 • Talcott Parsons 

the meeting of the various functional exigencies of the system, a 
new order of mechanisms of integration and of adaptation must 
emerge— the alternatives are dissolution of the system or its stabiliza- 
tion in an undifferentiated state. The theorem of primary signifi- 
cance here which, in the social system case, may ( following Durkheim 
again) be called the "law of organic solidarity" is as follows. In 
proportion as an action system becomes differentiated, there must 
be a balance between a) definiteness in the patterning of the rela- 
tions between units ( according to "rules" or norms in the normative 
reference) and b) flexibility yet consistency in the patterning of 
adaptation to the shifting exigencies of the situation, both of the 
system to its environment, and of the unit to its situation within 
the system. The "solution" of this dilemma lies in the institutional- 
ization of generalized normative patterns which are compatible with 
adaptive flexibility in particular situations; generalized legal systems 
are prototypical in this regard. 

These may be said to be the three fundamental dynamic theorems 
of the theory of action; the first is, in dynamic terms, a "law of 
inertia." At a next level down in the order of generality of functional 
significance I should place four additional propositions with refer- 
ence to the direct interchanges ( as distinguished from the diagonals 
in the paradigm) between the four primary functional subsystems 
of a system of action. Important special cases on the level of com- 
plex social systems which concern the relation of the economy to 
the rest of the society of which it is a part (of. Economy and 
Society, especially, Chap. II) are the balances involved in the com- 
modity and labor markets on the one hand, the capital markets 
on the other. A good example of the theorem element here is the 
assertion that the equilibrium of such a partial social system depends 
on a balance with reference to flows of labor power, commodities, 
and purchasing power; the Keynesian formulation of the immediate 
(as distinguished from the "monetary") conditions of maintenance 
of full employment is a classic example. 

It would lead much too far afield to enter here into even a 
sample of the many intricacies necessarily involved in presenting a 
complete account. It is clear that the logic behind these formula- 
tions is that of the processes of input-output interchange between 
subsystems of a system of action. The entities interchanged are. 


as noted above, classifiable as a) "resources," which are generated 
and consumed in the course of system process, and b) media, like 
money, which "circulate." The resources have the two fundamental 
"meanings" of facilities on the one hand, rewards on the other— 
another example of the duality principle running through the whole 

The general principle concerns on the one hand definition of 
the conditions of balance between flows in both directions which 
fall within the limits of maintenance of the relevant equilibria, and 
on the other identification of the consequences of exceeding those 
limits, in either direction (plus or minus) and on both sides of the 
interchange relationship. It goes without saying that for such gener- 
alizations to be formulated in terms which may be operationally 
testable, it is necessary to define the input and output categories as 
well as the relevant media and the conditional factors with sufficient 
clarity at the level of system performance and organization which 
is relevant to the particular problem. The fact that this is such a 
formidable task is, perhaps, even more than the primitive state of 
theory at the more general level, what makes it so difficult to come 
up with concrete operational criteria for testing. 

Generally, the most successful approach to progress in this task 
has, it seems to me, been achieved through codification of findings 
which were not originally sought in terms of hypotheses derived 
directly from this scheme. The case of certain economic market 
processes has already been noted. Similar cases at the macrosocial 
level are the balance between leadership, decision-making, and 
political support, ^^ and between social status elements and cultural 
patterns in the field of ideology. Direct independent research has, 
however, also played an important part.^^ 

A particularly good example involving the interchange between 
the two primary action systems of personality and social system 
is the process of socialization of the child, which enters into the 

16 Cf. Talcott Parsons, " 'Voting' and the Equilibrium of the Amer- 
ican Political System," in Eugene Burdick & Arthur Brodbeck ( eds. ) , 
American Voting Behavior, Glencoe, III.: The Free Press, 1959. 
i'^ Perhaps the best examples are Neil J. Smelser, Social Change in the 
Industrial Revolution, Chicago: Chicago Univ. Press, 1959, and 
Winston R. White, Ideology of the Intellectuals (Dissertation), Har- 
vard Univ., 1960. 

346 • Talcott Parsons 

papers of Baldwin and Bronfenbrenner. This is a case where the 
sociahzation system, e.g., of interaction between mother and child, 
is so structured as to produce a major disequilibrium in the per- 
sonality of the child. A new equilibrium can only be attained by a 
process of structural change in that system, one of the most impor- 
tant aspects of which is clearly diflFerentiation. 

One important type of operational application of some of these 
theorems, in which I have been recently and am currently engaged, 
concerns the interpretation of certain broad trends in the large-scale 
society, notably the American case. In a number of fields it has 
proved to be possible, by interpreting available data (on several 
levels) in relatively strict theoretical terms, to come to a clear 
choice between alternative interpretations of trends. Examples are 
( 1 ) trends in the American family; the hypothesis of diflFerentiation 
as distinguished from that of disorganization;^^ (2) trends in re- 
ligious organization; the hypothesis again of diflFerentiation as against 
"secularization" in the sense of decline in level of religious com- 
mitment; ^^ ( 3 ) trends in the relation of the individual to the group, 
in favor of the hypothesis of new levels of structural integration 
(of "institutionalized individualism") as against increased "con- 
formism" of the individual; ^^ and ( 4 ) conceptions of the positively 
integrative role of the mass media as opposed to the prevalent 
theory of the nature of "mass culture" in a "mass society." ^^ 

Perhaps the most important single case, however, is that of the 
problem of value-constancy vs. value-change in American society. 
Dr. Winston White and I have devoted very careful attention to this 
problem and have formulated the case for value-constancy with 
what we feel to be adequate theoretical clarity and empirical docu- 

^^ Cf. Family, Socialization and Interaction Process, Chapter I, Robert 
F. Bales, T. Parsons, James Olds, Morris Zelditch, & Philip E. Slater, 
Glencoe, 111.: The Free Press, 1955. 

1^ T. Parsons, "The Pattern of Religious Organization in the Con- 
temporary United States," Structure and Process in Modern Societies, 
Chap. 10, op. cit. 

2" T. Parsons and Winston White, "The Link Between Character and 
Society" in Continuities of Social Research, III, The Work of David 
Riesman, Glencoe, 111.: The Free Press, unpublished. 
21 Cf. T. Parsons and W. R. White, "The Mass Media & the Structure 
of American Society" in Journal of Social Issues, January, 1961. 


mentation. Our position on this problem stands in sharp contrast 
to what is probably the dominant current opinion of American social 
scientists. It could not have been arrived at and defended without 
a substantial element of technical theoretical analysis.-- S. M. Lipset, 
among others, takes a position similar to ours. 

I have organized a long discussion around Professor Black's con- 
tribution to the present volume because on the one hand, unlike 
that of Professor Devereux, it is critical rather than expository, and 
because among the critical papers, it is the one which raises in 
most direct form a number of crucial problems of the status of the 
theory of action as general theory. I am grateful to Black for the 
high level on which he has approached these problems, and for 
his clarity on a number of particular points on which there has 
been much misunderstanding, such as the illegitimacy of expecting 
immediate and easy translatability of general theory into opera- 
tional terms. At the same time it is quite clear that I cannot share 
his negative evaluation of the enterprise as a whole, including the 
opinion of its being premature in the present state of social science. 
I have pointed out that Black does not consider any developments 
of the scheme since the materials published in 1951. I do not 
think his verdict entirely fair for the level on which it stood at 
that time. But I also feel that there has been major progress since 
then, progress so important that it largely invalidates a judgment 
made as of the situation ten years ago. I fully recognize, however, 
the diflBculty in the position of a critic, not only because of the 
considerable volume of relevant writings and the fact that the 
position has in fact changed substantially— though I think in the 
direction of progressive revision and extension rather than of change 
of fundamental position. It has, I think, been a complex, but not 

22 In our opinion by far the most sophisticated exponent of the thesis 
of major value change has been the late Clyde Kluckhohn. ( Cf. "Have 
There Been Discernible Shifts in American Values During the Past 
Generation?" in The American Style, Elting Morison ( ed. ) , New 
York: Harper, 1958. Also "Shifts in American Values" in World 
Politics, Vol. XI, January 1959, No. 2.) My own statement is being 
reserved for a general book on American society. A partial statement 
of it will be published in the proceedings of the 1960 Conference on 
Science, Religion, and Philosophy under tlie title, "The Cultural Back- 
ground of American Rehgious Organization." 

348 • Talcott Parsons 

an erratic, course. Furthermore, there is still a considerable volume 
of unpublished materials— the most important recent published con- 
tribution having come too late to be used by any of the contrib- 
utors to the present volume— some in process of publication, but 
also a good deal not yet even written up in manuscript form but 
available only in notes, tabular classifications, and so forth, which 
only the author could decipher. With all these dijfficulties, however, 
I feel that more than the outline of a theoretical system of consider- 
able scope and power is now available. 

That its development is very far from complete goes without 
saying; if personal experience is any guide I should expect the 
changes which will come in the next decade to be as great as 
those of the last. Also the process continues to have the pragmatic 
character to which I referred in the beginning of this paper. It is 
in the nature of such a process that its products should need careful 
critical analysis and codification before a full verdict of their place 
in an attempt at developing general theory can be made. Only a 
few statements, like the "Pattern Variables Revisited" paper, can 
be direct statements of general theory as such. 

Let me now turn to a much briefer consideration of a few of the 
problems raised by the contributions to the volume other than 
Black's. At the outset it should perhaps be stated that my view of 
these papers is that they contain a complex combination of points 
on which there is excellent understanding and correct evaluation and 
some other points of serious misunderstanding. I should like to start 
with some of the points in the former category and divide them 
into what may be called procedural and substantive points. 

With respect to the former category may I express gratitude for 
Black's statement that this kind of enterprise requires a kind of 
intellectual daring. This well expresses my own feeling, and I take 
it not simply as a personal tribute, but as an expression of the view 
that social science needs more daring in the theoretical fields. I 
personally of course welcome evidence that more of it is developing, 
which indeed seems to be the case. Whatever the adequacies or 
inadequacies of my own contributions, on one point my conviction 


is unshakable, namely, that of the supreme importance of general 
theory in this field, as in others of science. We will not be scien- 
tifically mature until we construct such theory, and that cannot be 
achieved alone by empirical research and "letting the facts speak 
for themselves"; it can be achieved only through a great deal of 
hard, imaginative, frustrating and daring intellectual work at the 
level of theory itself. 

A second procedural consideration was indicated by Devereux 
which I also take as a tribute, and which points a similar moral; 
namely, that I have not made "disciples." I have been fortunate 
in being associated with a number of very able people— and in 
knowing the work of various others— who have made use of this 
conceptual scheme, have worked with it, have modified, extended 
and improved it and indeed at some junctures have gone off on 
tracks where I could not follow them. This seems to me to be 
entirely normal and desirable. A scientific theory is not something 
one "subscribes to" but which one uses, develops, tries out, modifies. 
I hope it is possible to keep the distinction clear between competent 
understanding of what has been done, which after all is the 
indispensable basis for good criticism positive or negative, and 
commitment to doctrines, the latter being inevitably grounded in 
noncognitive motives. I take it that it is the latter attitude which 
would characterize the disciple as Devereux means the term, and 
of these I want none. Indeed, the absence of such dependence 
( which is characteristically American ) seems to me to be an impor- 
tant index of scientific maturity. 

This relates to a third procedural point made by Professor Hacker, 
for which I am also grateful. There has of course been a close 
connection between the persistent criticism of "static bias" discussed 
in connection with Black's paper above, and an alleged "conserva- 
tive" bias in a political sense. The differentiation between scientific 
theorizing and the formulation of political attitudes and policies 
seems to me to be fundamental to the development of science, and 
particularly important in the social fields. But since in fact ideology 
plays into these things so much, it is important to have the record 
straight and part of this hinges in this case upon what is meant by 
the term "conservative." Hacker rightly points out that I personally 
have been, throughout my adult life as it happens, a relatively 

350 • Talcott Parsons 

typical American "egghead" intellectual, a New Dealer and Fair 
Dealer, and in general inclined to support the Democratic party, 
on its "liberal" side. In terms of the American political spectrum 
of course this is well to the left, and certainly not conservative in 
the sense in which a Robert Taft or a Herbert Hoover, or even an 
Eisenhower, have been conservatives. It is apparently too conserva- 
tive for Hacker, and of course for a good many other American 
intellectuals, especially those who think in more or less Marxist 
terms, but that is another issue. 

Both Devereux and Williams, as I have already noted, are very 
straightforward and unequivocal in presenting what, to me, is the 
correct interpretation of the significance of and attitude toward 
the problem of order in my work. This of course is that the basis 
of social order is inherently problematical and, in the nature of the 
case, had to form one of the major foci of preoccupation for socio- 
logical theory, as was particularly true for the group of European 
theorists treated in the Structure of Social Action. This applies most 
explicitly to the position of Durkheim. To interpret this concern 
for order, as a theoretical problem, as justifying the allegation of 
a bias in favor either of static problems over dynamic, or of con- 
formity over originality or creativity is, as I have stated, a gross 
distortion. I am indeed grateful to these two contributors for setting 
the record straight on this very fundamental point. 

The sociological solution of the problem of order is, as I have 
stated above, an analysis of the part played in social systems by 
institutionalized patterns of normative culture. Institutionalization 
here clearly involves more than the authority of Hobbes' sovereign; 
it implies, so far as authority is concerned, legitimation in Weber's 
sense, and a whole range of mechanisms involving the maintenance 
and operation of values and norms in all spheres of social life, not 
only the political. Institutionalization, however, would not be pos- 
sible if the relation of the institutional structure of the society to the 
other structures and to the "interests" of individuals were either 
random or were based only on the coercive type of control through 
negative sanctions. Professors Morse, Baldwin, and Landsberger 
all present very fundamental insights into diflFerent phases of this 
essential point. 

In the case of Morse it concerns the problem of the social environ- 


ment of the economy, taken as a differentiated subsystem of a com- 
plex society. Of course in some sense every economist has been 
aware that there was such an environment, since after all economic 
theory is not a theory of society as a whole, and hence aware that 
in some sense this environment was structured. At the same time 
it has been a glaring gap in the main traditions of economic theory 
that scarcely any serious attempt has been made to tackle this 
problem in genuinely theoretical terms. On the contrary, the typical 
procedure of economists has been to follow through the logic of 
economic theory as far as it would take them and then to jump 
to a series of ad hoc common sense qualifications of the economic 
scheme, knowing, of course, that it does not take adequate account 
of noneconomic factors. 

The tacit assumption is that the noneconomic factors cannot be 
analyzed on a level of theoretical specificity comparable to that of 
economic theory and, even if they could, no theoretically specific 
articulation between the economic and the noneconomic schemes 
can be established.^^ 

It is precisely as an attempt to get out of this scientifically im- 
possible impasse, that I should like to have the work of Smelser 
and myself on this problem judged. A partial, and hence far from 
satisfactory, approach to it was worked out years ago in the Structure 
of Social Action. Economy and Society, however, represents in my 
opinion a very substantial advance. 

The essential point there is the treatment of the historic economic 
classifications of the factors of production— land, labor, capital and 
organization— and the corresponding shares of income, as special 
cases of the categories of input and output respectively, interpreted 
as the inputs and outputs of a subsystem of a total society from and 
to the other subsystems. Given this orientation it is then possible to 
interpret the sources and destinations of these inputs and outputs as 

23 Though it is an old book, Lionel Robbins' Nature and Significance 
of Economics has never been thoroughly transcended within the main 
traditions of economics, in its exphcit position that the "environment" 
of economic behavior must be treated as theoretically random. Cf. 
my critique, "Some Reflections on The Nature and Significance of 
Economics," Quarterly Journal of Economics, XLVIII, pp. 511-545, 

352 • Talcott Parsons 

theoretically specific; thus, for instance, the labor-consumption 
boundary of the economy is to be interpreted as standing vis-a-vis the 
pattern-maintenance subsystem of the society, which includes the 
primary functional reference of the family household. 

If such an interchange is one operating between a subsystem with 
primacy of economic, i.e., in the technical economic sense "pro- 
ductive" function, and one of primarily noneconomic function in 
the society, then the stability of the interchange must, on the prin- 
ciples stated above, be dependent on a balance of the conditions 
important on both sides. Thus, in interchanges between firms and 
households, the economic primacy of the firm, which, in a differ- 
entiated economy, is organized about the criterion of solvency in 
the relevant sense, must be balanced with the consumption tastes 
and the security interests of households. To be sure, households 
also are typically subject to the standard of solvency in the sense 
that the money expenses of the operation of the normal household 
are expected to be covered by the money income of its members. 
But in the case of the firm it is its productive efficiency which must 
be matched against the conditions of solvency, v/hereas in that of 
the household it is what might be called the integrity of its pattern 
of fife. 

The most essential point that Morse makes is, as I understand it, 
that the historic pattern of economic theory really makes no pro- 
vision for the analysis of this essential balancing; it tends to treat 
every interchange which is in any sense economic, as either 
"governed" by principles of economic rationality or "deviant" in 
this respect or some combination of these. Hence, in the fields men- 
tioned for illustration, the difficulty exists of a satisfactory theoretical 
treatment of the problems of consumption and standards of living 
on the one hand, and of the structure of the labor market, the role of 
trade unions, and so forth, on the other. 

It would follow from the position Smelser and I have taken that 
the nearest approaches to the theoretical economist's conception of 
theoretical fit would be found in interchanges internal to the econ- 
omy and that, as the interchanges over its boundaries are ap- 
proached, progressively greater modifications of these theoretical 
patterns become necessary for adequate interpretation. It is his 
clear insight into this crucial point which, to my mind, is the great 


merit of Morse's analysis, and with it the reaHzation that, to reach 
a higher level of theoretical generalitv^ economic theorv^ must be 
substantially modified to make possible its theoretical integration 
wdth the other branches of the theorv of social svstems. 

In a very different empirical field, Baldwin emphasizes a different 
special case of the same basic theoretical point. He calls attention 
to the fact that psychologists very generally have failed to give 
explicit consideration to the structure of the situation in which the 
personalities of individuals function. They have, rather, tended to 
treat stimuli as given in the situation, and hence in principle dis- 
crete, and then to try to trace the consequences for the individual 
of exposure to different classes of stimuH in different states of the 
personality system. He then speaks of my own contributions in 
this field as important in placing a very strong emphasis on the 
structure of the situation. 

Tliis is correct, and a most welcome recognition. For the human 
personahty the particularly important point is that the "stimulus 
situation" of primary importance is the social and cultural environ- 
ment in which it functions. This is crucial from the earliest begin- 
nings of what we call the process of socialization. It is a point on 
which I think Freud had the basically correct insight, especially in 
the sense in which his "reahty principle" in practice, in the field 
of object relations, was interpreted to mean the reality of social 
interaction. But this means that, in turn, the structure of that 
environment is the structure of the nesting series of social and 
cultural systems in which the individual is placed and that, if it 
is to be taken into account in a technically theoretical sense, it 
must be through their analysis as such systems in their own right, 
e.g., for the case of socialization starting with the nuclear family.-* 

The theory of action, which treats social systems and personality 
systems as subsystems of the same more general system of action, 
is in my opinion better equipped to carry through the implications 

2^ I am aware that, in Family, Socialization . . . op. cit., the analysis 
of the postoedipal phases of sociafization and the social structure into 
which they fit was sketchy and incomplete. This gap has been par- 
tially fiUed in two subsequent papers; namely, Parsons & White, "The 
Link Between Character . . ," op. cit., and "The School Class as a 
Social System," Harvard Educational Review, Fall, 1959. 

354 • Talcott Parsons 

of this essential insight than previous conceptual schemes. I hope 
it is clear to the reader that the nature o£ the theoretical problem 
is directly parallel to that of the boundaries of the economy just 
discussed. The tendency has been to treat the theory of personality 
as an entirely independent scheme which was not in any specific 
sense theoretically articulated with either social or cultural system 
theory, though in the latter case the "culture and personality" 
school have made certain (to me) rather unsatisfactory attempts. 
But if personality is in fact a subsystem of a more comprehensive 
system, then for it to be satisfactorily analyzed, it is necessary to 
work out its specific articulations with the other subsystems on 
which it impinges and with which it carries on input-output inter- 
changes. Technical consideration of the structure of its environment 
is the first essential step in such a program, but in my opinion 
only the first step. Certain further ones lead into the problem 
Baldwin refers to as that of the "isomorphism" between social sys- 
tems and personalities, on which I will say a word presently. 

When it comes to the treatment of formal organizations as sub- 
systems of a society the same order of problem arises. Especially 
with the prominence accorded to the problems of the relation 
between formal and informal organization in the more sociological 
area of American work in this field, it is natural that primary 
emphasis should have been placed on the internal problems of 
organizations such as the line-staff problem, the clique problem 
and various others. The logic of this tendency is exactly like that of 
personality psychology; the social environment of the organization 
has tended to be treated as unproblematical (a position standing 
in extreme antithesis to the "pure" economic theory of the firm as 
only a pinpoint at which money income and expenditure are in 
balance— or not in balance, as the case may be.) It is, in my view, 
the greatest merit of Landsberger's paper that its author has 
emphasized the importance given to the social environment of the 
organization as one major basis on which analysis of the internal 
differentiation of structure and functions has to be approached. 
There are, both in his and in Whyte's paper, complaints about my 
not having said enough concerning internal function or organization. 
But the critical point is recognition of the importance of dealing 
with this neglected problem area, and of the place of the problems 


<ioncerned in the larger theoretical framework. What I have in 
mind here is the distinction between what I have called the tech- 
nical, the managerial and the "institutional" or fiduciary components 
in the functions of an organization and the ways in which these 
distinctions play into the organization's relations to its environment.^^ 

It can thus, I hope, be seen that the three problems have in com- 
mon their reference to a certain "atomism" of theoretical tradition 
in the social sciences, an element which I interpret to be a heritage 
from utilitarianism. The same problems arise again at the level of 
^'behavior" psychology in the Hullian tradition of focusing on 
discrete "stimulus-response" units and at various other points. The 
countervailing emphasis on the structure of environments is directly 
connected with that on levels of organization and their relation to 
cybernetic control, which has come up again and again. In this 
connection, then, it can be seen to be an essential point that in the 
environments of each of the three systems dealt with in these papers, 
there is a set of components which is hierarchically super ordinate 
to the system of reference. 

Though what on various occasions I have called the polity stands 
in this relation to the economy, the case of greatest sociological 
interest is the integrative subsystem of the society with its frame- 
work of institutionalized norms. In the case of the economy this 
includes above all the institutional complexes having to do with 
rights in possessions and with the structure of markets; namely, 
contract, employment and property ( cf. Economy and Society, 
Chap. Ill), but of course others, notably for the firm as an organi- 
zation, authority. 

For the case of the personality system it is crucial that in the 
general hierarchy of action both social and cultural systems are 
superordinate to personalities; in a sense which must be very care- 
fully interpreted— it is emphatically not a sense which is incom- 
patible with the value— complex of individualism. Similarly, the 
environment of formal organizations includes, of course, access to 
facilities and channels of disposal, but it also includes an institu- 

25 Cf. "Some Ingredients . . . ," Structure 6- Process in Modern So- 
cieties, op. cit., Ghap. II; and Robert K. Merton, Leonard Broom, & 
Leonard S. Cottrell, Jr., (eds.), Sociology Today, Chap. I, New 
York: Basic Books, Inc., 1959. 

356 • Talcott Parsons 

tionalized framework of norms and values, which above all impinge 
on the organization through what I have called the institutional 
or fiduciary complex. 

It therefore seems to me to be justified to assert that the concept 
of the institutionalization and/or internalization of the structure of 
superordinate systems— in the specific sense of the hierarchy of con- 
trol of action systems— is one crucial common feature of the relation 
of every system or subsystem of action to its environment.^^ 

This relationship is the basis of the very fundamental proposition, 
stated above, that the structure of systems of action consists in insti- 
tutionalized and internalized patterns of normative culture. Mean- 
ing, in the cultural sense, is the master category of the structure of 
systems of action, a proposition which is expressed in the placing 
of the cultural system at the top of the cybernetic hierarchy for 
action as a whole. Then the basic condition of the integration of 
systems of action in this hierarchical direction is that each system 
in the series should spell out or "actualize," for its specified condi- 
tions, the implications of the meanings institutionalized ( or internal- 
ized) in the next-higher system. For the case of social systems we 
refer to this as the specification of the cultural values to the level 
of the functioning of a social system. 

The problem of the "isomorphism" of personality and social sys- 
tems, which bothers Baldwin so seriously, is a special case of this 
general problem of action systems. To me the term "isomorphism" is 
a dubious one, because it seems to me to claim too much. What I 
contend is that an essential set of components are common to 
systems which, like personality and social system, are adjacent in 
the hierarchical series. This is the problem which, on various 
occasions,^^ I have formulated as that of the interpenetration 
between systems. It involves an element of common structure 
which, in its own genetic history, has been acquired by the lower- 
order system in the specific sense of this discussion— in the course 
of interaction with the higher-order system, the prototypical case 
being socialization of the individual. In this sense the main outline 
of the structure of human personalities consists in the social object 

26 What I call the cultural system is clearly a limiting case in this 
respect. Cf. "Culture and the Social System," op. cit. 

27 E.g., "An Approach to Psychological Theory . . . ," op. cit. 


systems which have been internalized in the course of the indi- 
vidual's life history. I am aware that this is a radical proposition, 
and that it is unacceptable to Baldwin, but it seems to me to 
be essential.^^ 

The other side of the question, which bears with it the reason 
why the term "isomorphism" is a dubious one, is that each of these 
two systems, like any differentiated pair, is grounded in different 
exigencies, so that, given certain of the some structural pattern 
components, the empirical and dynamic problems are typically 
different in each case. In the case at hand, social systems are 
grounded, at this level, in the exigencies of the interaction of 
pluralities of persons. Their goals must ultimately be oriented to 
the motivational needs and interests of individuals; their adaptive 
interests must be concerned with the environment which is common 
to individuals and groups, not the least important aspect of which 
is the cultural, and their integrative problems are by definition 
interpersonal. For personalities, on the other hand, the exigencies 
are quite different. Their goals must be organized about the 
structure of the social situation of the individual, the obverse of the 
case for the person. The cultural tradition is, to be sure, an environ- 
mental element common to the two, though in different contexts. 
But their integrative exigencies are intrapersonal. Finally, the per- 
sonality of the individual has a special direct relation to the living 
organism which, in the case of the social system, is indirect rather 
than direct. This relation has, in my opinion, been most forcefully 
formulated, in modern psychology, by Freud in the concept of the Id 
and its relation to the pleasure principle. Whatever the "biological" 
factor in the concrete functioning of societies, it is clearly not an 
Id in the Freudian sense. 

In Baldwin's argument I sense an anxiety that personality theory 
will somehow be "reduced" to social system terms and the sociol- 
ogists will then "take over." The anxiety is surely groundless. To be 

28 In my own case it has been most fully developed in the paper 
"Social Structure and the Development of Personality," Psychiatry, 
November, 1958. To me it is rather surprising that neither Baldwin 
nor Bronfenbrenner seems to take account of what I feel to be the 
important advances documented in this paper over the position taken 
in Family, Socialization . . . , op. cit. 

358 • Talcott Parsons 

sure, Durkheim, under the stress of insisting that the "social" com- 
ponent could not be neglected, sometimes seemed to assert that 
what was not social was "purely" biological. But we have gone 
beyond Durkheim in this respect. Baldwin himself is very clear 
that much of psychological tradition has grossly neglected the social 
component. Once this historic imbalance— and I claim strenuously 
that it has, in the main traditions of psychology, been an imbalance 
—is righted, I see no fundamental difficulty in arriving at a broad 
definition of the relations of the components which should prove 
to be acceptable to the respective professional groups. And more 
important, this definition can provide a basis on which not only 
each group can go about its business, but also the increasingly impor- 
tant area of articulation and interpenetration can be investigated 
without bogging down in what are, at the professional level, essen- 
tially ideological controversies. At any rate, as I see it, the sphere 
left for the theorist of personality within the psychological tradition 
is not only not unduly restricted by the theory of action, but the 
range of baffling and challenging problems is immense. Sociologists 
are going to have quite enough to do in other fields without attempt- 
ing to usurp this one. 

It should go without saying that the same principles apply to 
other cases of the interpenetration of adjacent systems as to that 
of social system and personality. In my opinion the fact of inter- 
penetration is crucial, but so also is that of the independence of the 
systems from each other in the analytical sense of independence. 
Like the famous case of heredity environment, it is not a case of 
one or the other, but of understanding of the complex ways in 
which both are involved and interact. 

There is a considerable range of further problems discussed in 
the various contributions to this volume which, if there were space 
available, it would be fruitful to take up. As already noted, however , 
it has seemed to me more important to use the space available 
mainly to attempt to clarify some of the central problems of the status 
of the action scheme at the level of general theory, even at the 
risk of relative neglect of a number of the more specific problems 
which have loomed largest in the minds of a number of contributors. 
The main justifications of this pohcy seem to me to be two. The 
first is that, as a theorist in this field, I have laid a particularly 


strong stress on the importance of general theory and invested a 
larger part of my own eflForts in this direction than others have 
and hence am perhaps especially qualified to speak on this level. 
Second, it seems to me that if some of the basic questions which 
have been reviewed here can be cleared up at the level of general 
theory it should help enormously to clarify the problems which 
have to be faced in the "middle ranges" by providing criteria for 
approaching many of these problems. I have tried to illustrate 
this in the last few pages by taking up the relation of the problem 
of the structure of environments of subsystems to the general prob- 
lem of order which is such a central over-all theme and to the 
status of the hierarchy of control. I hope I have been able to con- 
vince the reader that, precisely on the level of general theory, the 
problem is essentially the same for the economy in relation to the 
institutional order of the society, for the personality in relation to 
the social object systems, and for the formal organization in rela- 
tion to its social environment. Morse, Baldwin, and Landsberger 
tend to treat these as discrete problems and of course as such they 
are the fully legitimate concern of specialists in each field. But in 
the present context the common element is of crucial significance, 
since it is only by demonstrating the presence and importance of 
such common elements that the kind of codification of these many 
fields under a single theoretically integrated conceptual scheme can 
be worked out. 

In the meantime I am acutely aware that a number of authors 
are unsatisfied on a number of issues; e.g., Whyte on "where is the 
action in action?", Bronfenbrenner on the psychology of identifica- 
tion, particularly at specific operational levels, and Hacker on the 
place and amount of conflict in American society. My only excuse 
for not going farther toward attempting to satisfy them, and of 
course others on other points, is that, within the limits of a single 
rather long essay, it is not possible to do everything which might 
be relevant. My selection has been made in the light of my own 
judgment of the most strategic issues, but of course this judgment 
may well be fallible. In any case, this seems to me to be an excellent 
example of the difficulties of integration of such a variegated field 
as that involved in the present volume. It makes it quite clear that 
attainment of the ultimate goal is still very far ahead, that goal 

360 • Talcott Parsons 

being a unified conceptual scheme which is taken for granted in 
the relevant professions and, as a matter of course, used as the bas^ 
of operations for exploring the problems on the new frontiers oi 
knowledge. Whatever the final judgment on my own contributions, 
I am deeply convinced that the ultimate development of such a 
scheme is essential and that the general trend of intellectual history 
is in that direction. 

In conclusion perhaps I may attempt to ground the above con- 
viction by very briefly raising a few questions about this major 
trend. I have suggested above that the "modern" level of sociologi- 
cal thinking has emerged from the intellectual "marriage" between 
the utilitarian and the more coUectivistic elements in the main 
Western traditions of social thought. This general theme can, I 
think, be generalized to the treatment of action as a whole, indeed 
even more broadly to the phenomena of life if not even to include 
the physical world. 

It seems to me to follow from this view that neither old-style 
"atomism," nor in certain meainings of the term "individualism," 
nor old-style "idealism," is tenable any more. Because of certain 
features of American cultural history there is little probability that 
idealistic theories will gain ascendancy, though there has been a 
definite strain of this sort in much American anthropology in the 
tradition of Boas. There are, however, three major cases of the 
atomistic orientation which have, played a very important role. 
These are, first, the theory of "economic individualism," including 
its generalization in a Spencerian type of individualistic rationalism; 
second, what may be called personality-focused individualism, with 
its overwhelming suspicion of any independent integrative signifi- 
cance attributed to society or culture— Floyd Allport's Institutional 
Behavior was a major document of this view of a generation ago; 
and third, what may be called "behavioristic atomism" of which the 
hypostatization by Hull and his followers of the "S-R" unit is the 
prime example. 

It is my contention that this old-style utilitarian atomism is no 


longer tenable on any of these levels; as I have noted, it cannot 
account for the fundamental phenomena of organization. To put it 
very schematically, on the level of the behavioral organism this 
view has been made untenable by the work of the "ecologists," by 
the Tolman type of theory, and not least by recent work on brain 
functioning. On the level of the personality it is made untenable 
by Freud above all, as well as by Kurt Lewin, Murray, Ericson, and 
others. On the level of social systems it can be said that its demise 
was clearly foreshadowed in 1848, the year in which the last great 
utilitarian document— John Stuart Mill's Principles of Political 
Econo7mj— coincided in publication with the Communist Manifesto. 

It is no accident that the problem of socialism was not only 
politically but also intellectually so salient in the late 19th century in 
the West. There seems to me to be no doubt that, along with various 
others, Marx was perhaps the most important original "go-between" 
of the marriage of the two basic traditions we are discussing. But 
the marriage was consummated, not in the work of Marx, but in 
the work of a generation following him, of which the critical names 
are Durkheim, Weber, and Freud. The very fact that all three 
were more scientists than they were political figures has obscured 
their basic significance relative to Marx. 

I am very happy to acknowledge that, on the sociological level, 
Marx is one of the symbolic "grandfathers" of the theory of action. 
Marx, however, as Schumpeter made so clear, (Capitalism, Socialism 
and Democracy) had a dual personality, as social scientist and as 
revolutionary leader. The first condition of his fathering a science 
was the diflFerentiation of the two components, a process which, by 
and large, did not occur until the next generation. In spite of 
ambivalences on this point, the three great figures just named were, 
in a sense in which this was not true for Marx, in the first instance 

To me, what I call the theory of action was, in its core— which 
I take to be the social system in its relation to the personality of 
the individual— founded in the generation of Durkheim, Weber, and 
Freud, with of course a very complex set of other influences, a 
few of which have been mentioned here. With the very perceptible 
fading of the influence of the older economic individualism— in its 
scientific rather than political reference— and the older personaHty- 

362 • Talcott Parsons 

individualism, it is a striking fact that general orientations in this 
field have, in recent years, tended increasingly to polarize between 
a nondogmatic and nonpolitical "Marxian" position and one which 
in the broadest sense may be called one or another version of the 
theory of action, ^^ The most important exception to this is probably 
the influence of the "culture and personality" school which is an 
attempted direct fusion of the atomistic and idealistic trends,^*^ as 
distinguished from what I have symbolized as a "marriage." 

Seen in the perspective of recent intellectual history, it seems to 
me to be an inescapable conclusion that the relation between the 
Marxian type of thought and that represented by the turn of the 
century generation of action theorists is one of sequence in a 
definite developmental trend. Relative to its antecedents, clearly 
the Marxian trend is an advance at the level of general theory. It 
is not, for my purposes however, important so much as a "mate- 
rialistic" interpretation, as it is as a less differentiated basis of 
analysis of the great problems of cultures, personalities, and social 
systems as they arise for social scientists in our day. 

This is the main perspective in which I hope evaluation of the 
conceptual scheme discussed here as the "theory of action" will 
settle down. It is my own profound conviction, which I both believe 
and hope to be justified, that the developments under discussion in 
this volume are deeply rooted in the main trends of the intellectual 
development of our age. Their base-line of reference is a great 
synthesis which was achieved in the generation preceding ours— as 
the editors of Theories of Society have placed it, roughly 1890-1935. 
This synthesis has provided an essentially new orientation for 
general theorizing over a wide range of concerns with human 

29 C/. S. M. Lipset, review of B. Moore, Jr., Political Potver and 
Social Theory, in the American Sociological Review, April, 1960. 
3*^ Full documentation of this judgment would require an extended 
essay in intellectual history. The most striking evidence is the curious 
aflBnity between Boasian "culture" theory, from configuration version 
to trait version, with Hullian learning theory, via a very specially 
selective use of psychoanalytic theory centering on the alleged 
specificities of the effects of very particular child training practices. 
The similarity of the logical structure of this body of thought with 
that of German idealistic thought of the late 19th century is striking 
indeed. Cf. Talcott Parsons, Structure of Social Action, Chaps. XIII 
and XVI, New York: McGraw-Hill, 1937. 


behavior, but also with a new emphasis, even in the case of Freud, 
on the social dimensions of the organization of behavior. 

Development from this base-line must, as I see it, follow some- 
thing of the type of pragmatic pattern I have outlined in the 
first main section of this paper. For it even to attempt to jump 
directly to the deductive system would be entirely out of tune 
with the main spirit of the great tradition in which this scheme 
purports to have a place. But within this framework, it attempts 
to make theory better integrated, more precise, more explicit, more 
wide-ranging in its empirical references. As an essential part of this 
task it attempts as systematically as possible to codify empirical 
knowledge and to integrate the empirical generalizations which 
can be related to it directly with the propositions of general theory. 

If it is judged that this has in fact been an authentic tradition of 
modern intellectual development, and that the theory of action 
has made some important contribution to clarification, generaliza- 
tion, and codification within it, it should follow that everything con- 
sidered we have made a step toward better scientific understanding 
of the human condition as a result of this work. Such a judgment 
would be an ample reward for whatever efforts have gone into 
it. The work of the Cornell group, as documented in this volume, 
has, at the very least, served to bring the problems of whether 
such a judgment is justified much more sharply into focus than 
would have been possible without it. 


\ ■ V 



\ Date Due 
Due \ Returned Due Returned